@@ -1,17 +1,24 @@
.. SPDX-License-Identifier: BSD-3-Clause
Copyright(c) 2022 Intel Corporation
-=======================================================================
+=====================================================================
Loopback vhost-user/virtio-user server mode with DSA driver test plan
-=======================================================================
+=====================================================================
Description
===========
Vhost asynchronous data path leverages DMA devices to offload memory copies from the CPU and it is implemented in an asynchronous way.
-In addition, vhost supports M:N mapping between vrings and DMA virtual channels. Specifically, one vring can use multiple different DMA
-channels and one DMA channel can be shared by multiple vrings at the same time. Vhost enqueue and dequeue operation with CBDMA channels is supported
-in both split and packed ring.
+Asynchronous data path is enabled per tx/rx queue, and users need to specify the DMA device used by the tx/rx queue.
+Each tx/rx queue only supports to use one DMA device, but one DMA device can be shared among multiple tx/rx queues of d
+ifferent vhostpmd ports.
+
+Two PMD parameters are added:
+- dmas: specify the used DMA device for a tx/rx queue.(Default: no queues enable asynchronous data path)
+- dma-ring-size: DMA ring size.(Default: 4096).
+
+Here is an example:
+--vdev 'eth_vhost0,iface=./s0,dmas=[txq0@0000:00.01.0;rxq0@0000:00.01.1],dma-ring-size=4096'
This document provides the test plan for testing the following features when Vhost-user using asynchronous data path with
DSA driver (kernel IDXD driver and DPDK vfio-pci driver) in loopback virtio-user topology.
@@ -29,7 +36,7 @@ If iommu on, kernel dsa driver only can work with iova=va by program IOMMU, can'
Note:
1. When DMA devices are bound to vfio driver, VA mode is the default and recommended. For PA mode, page by page mapping may
exceed IOMMU's max capability, better to use 1G guest hugepage.
-2. DPDK local patch that about vhost pmd is needed when testing Vhost asynchronous data path with testpmd, and the suite has not yet been automated.
+2.DPDK local patch that about vhost pmd is needed when testing Vhost asynchronous data path with testpmd.
Prerequisites
=============
@@ -44,8 +51,8 @@ General set up
# CC=gcc meson --werror -Denable_kmods=True -Dlibdir=lib -Dexamples=all --default-library=static <dpdk build dir>
# ninja -C <dpdk build dir> -j 110
For example:
- CC=gcc meson --werror -Denable_kmods=True -Dlibdir=lib -Dexamples=all --default-library=static x86_64-native-linuxapp-gcc
- ninja -C x86_64-native-linuxapp-gcc -j 110
+ CC=gcc meson --werror -Denable_kmods=True -Dlibdir=lib -Dexamples=all --default-library=static x86_64-native-linuxapp-gcc
+ ninja -C x86_64-native-linuxapp-gcc -j 110
2. Get the PCI device ID and DSA device ID of DUT, for example, 0000:4f:00.1 is PCI device ID, 0000:6a:01.0 - 0000:f6:01.0 are DSA device IDs::
@@ -106,9 +113,9 @@ Common steps
Check WQ by 'ls /dev/dsa' and can find "wq0.0 wq1.0 wq1.1 wq1.2 wq1.3"
Test Case 1: Loopback split ring server mode large chain packets stress test with dsa dpdk driver
----------------------------------------------------------------------------------------------------
+-------------------------------------------------------------------------------------------------
This is a stress test case about forwading large chain packets in loopback vhost-user/virtio-user split ring with server mode
-when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver. Both iova as VA and PA mode test.
+when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
1. Bind 1 dsa device to vfio-pci like common step 1::
@@ -117,8 +124,8 @@ when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk dr
2. Launch vhost::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -n 4 -l 2-3 --file-prefix=vhost -a 0000:e7:01.0,max_queues=1 \
- --vdev 'net_vhost0,iface=vhost-net0,queues=1,dmas=[txq0;rxq0],client=1' \
- --iova=va -- -i --nb-cores=1 --txd=1024 --rxd=1024 --mbuf-size=45535 --lcore-dma=[lcore3@0000:e7:01.0-q0]
+ --vdev 'net_vhost0,iface=vhost-net0,queues=1,dmas=[txq0@0000:e7:01.0-q0;rxq0@0000:e7:01.0-q0],client=1' \
+ --iova=va -- -i --nb-cores=1 --txd=1024 --rxd=1024 --mbuf-size=65535
3. launch virtio and start testpmd::
@@ -129,59 +136,43 @@ when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk dr
4. Send large pkts from vhost and check the stats::
- testpmd>set txpkts 45535,45535,45535,45535,45535
+ testpmd>set txpkts 65535,65535
testpmd>start tx_first 32
testpmd>show port stats all
-5. Stop and quit vhost testpmd and relaunch vhost with pa mode by below command::
-
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -n 4 -l 2-3 --file-prefix=vhost -a 0000:e7:01.0,max_queues=4 \
- --vdev 'net_vhost0,iface=vhost-net0,queues=1,dmas=[txq0;rxq0],client=1' \
- --iova=pa -- -i --nb-cores=1 --txd=1024 --rxd=1024 --mbuf-size=45535 --lcore-dma=[lcore3@0000:e7:01.0-q0]
-
-6. Rerun step 4.
-
Test Case 2: Loopback packed ring server mode large chain packets stress test with dsa dpdk driver
-----------------------------------------------------------------------------------------------------
+--------------------------------------------------------------------------------------------------
This is a stress test case about forwading large chain packets in loopback vhost-user/virtio-user packed ring with server mode
-when vhost uses the asynchronous operations with dsa dpdk driver. Both iova as VA and PA mode test.
+when vhost uses the asynchronous operations with dsa dpdk driver.
1. Bind 1 dsa port to vfio-pci as common step 1::
- <dpdk dir># ./usertools/dpdk-devbind.py -b vfio-pci f1:01.0
+ <dpdk dir># ./usertools/dpdk-devbind.py -b vfio-pci e7:01.0
2. Launch vhost::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-3 -n 4 -a 0000:f1:01.0,max_queues=1 \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=1,dmas=[txq0;rxq0],client=1' \
- --iova=va -- -i --nb-cores=1 --mbuf-size=45535 --lcore-dma=[lcore3@0000:f1:01.0-q0]
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=1,dmas=[txq0@0000:e7:01.0-q0;rxq0@0000:e7:01.0-q0],client=1' \
+ --iova=va -- -i --nb-cores=1 --mbuf-size=65535
3. launch virtio and start testpmd::
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 5-6 -n 4 --file-prefix=testpmd0 --no-pci \
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 5-6 -n 4 --file-prefix=testpmd0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=1,mrg_rxbuf=1,in_order=0,vectorized=1,packed_vq=1,queue_size=2048,server=1 \
-- -i --rxq=1 --txq=1 --txd=2048 --rxd=2048 --nb-cores=1
testpmd>start
4. Send large pkts from vhost and check the stats::
- testpmd>set txpkts 45535,45535,45535,45535,45535
+ testpmd>set txpkts 65535,65535
testpmd>start tx_first 32
testpmd>show port stats all
-5. Stop and quit vhost testpmd and relaunch vhost with pa mode by below command::
-
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-3 -n 4 -a 0000:f1:01.0,max_queues=1 \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=1,dmas=[txq0;rxq0],client=1' \
- --iova=va -- -i --nb-cores=1 --mbuf-size=45535 --lcore-dma=[lcore3@0000:f1:01.0-q0]
-
-6. Rerun step 4.
-
-Test Case 3: Loopback split ring all path server mode and multi-queues payload check with dsa dpdk driver
------------------------------------------------------------------------------------------------------------
+Test Case 3: Loopback split ring inorder mergeable path multi-queues payload check with server mode and dsa dpdk driver
+-----------------------------------------------------------------------------------------------------------------------
This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split ring
-all path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
-Both iova as VA and PA mode test.
+inorder mergeable path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
+
1. bind 3 dsa port to vfio-pci like common step 1::
@@ -191,12 +182,11 @@ Both iova as VA and PA mode test.
2. Launch vhost::
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=4 \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0;txq1;txq2;txq3;txq4;txq5;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore11@0000:e7:01.0-q0,lcore12@0000:e7:01.0-q0,lcore13@0000:e7:01.0-q1,lcore13@0000:e7:01.0-q2,lcore14@0000:e7:01.0-q1,lcore14@0000:e7:01.0-q2]
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=8 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-3. Launch virtio-user with split ring mergeable inorder path::
+3. Launch virtio-user with split ring inorder mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=1,server=1 \
@@ -221,9 +211,25 @@ Both iova as VA and PA mode test.
6. Quit pdump, check all the packets length are 6192 Byte in the pcap file and the payload in receive packets are same.
-7. Quit and relaunch vhost and rerun step 4-6.
+Test Case 4: Loopback split ring mergeable path multi-queues payload check with server mode and dsa dpdk driver
+---------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split ring
+mergeable path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
+
+
+1. bind 3 dsa port to vfio-pci like common step 1::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u e7:01.0 ec:01.0 f1:01.0
+ ./usertools/dpdk-devbind.py -b vfio-pci e7:01.0 ec:01.0 f1:01.0
+
+2. Launch vhost::
-8. Quit and relaunch virtio with split ring mergeable path as below::
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=8 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+3. Launch virtio-user with split ring mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=0,server=1 \
@@ -231,9 +237,42 @@ Both iova as VA and PA mode test.
testpmd>set fwd csum
testpmd>start
-9. Rerun steps 4-7.
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,64,64,2000,2000,2000
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 6192 Byte in the pcap file and the payload in receive packets are same.
+
+Test Case 5: Loopback split ring non-mergeable path multi-queues payload check with server mode and dsa dpdk driver
+-------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split ring
+non-mergeable path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
+
-10. Quit and relaunch virtio with split ring non-mergeable path as below::
+1. bind 3 dsa port to vfio-pci like common step 1::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u e7:01.0 ec:01.0 f1:01.0
+ ./usertools/dpdk-devbind.py -b vfio-pci e7:01.0 ec:01.0 f1:01.0
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=8 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+3. Launch virtio-user with split ring non-mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=0,server=1 \
@@ -241,9 +280,13 @@ Both iova as VA and PA mode test.
testpmd>set fwd csum
testpmd>start
-11. Rerun step 4.
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
-12. Send pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
testpmd> set fwd csum
testpmd> set txpkts 64,128,256,512
@@ -252,11 +295,27 @@ Both iova as VA and PA mode test.
testpmd> show port stats all
testpmd> stop
-13. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+
+Test Case 6: Loopback split ring inorder non-mergeable path multi-queues payload check with server mode and dsa dpdk driver
+---------------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split ring
+inorder non-mergeable path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
+
+
+1. bind 3 dsa port to vfio-pci like common step 1::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u e7:01.0 ec:01.0 f1:01.0
+ ./usertools/dpdk-devbind.py -b vfio-pci e7:01.0 ec:01.0 f1:01.0
-14. Quit and relaunch vhost and rerun step 11-13.
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=8 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-15. Quit and relaunch virtio with split ring inorder non-mergeable path as below::
+3. Launch virtio-user with split ring inorder non-mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=1,server=1 \
@@ -264,9 +323,42 @@ Both iova as VA and PA mode test.
testpmd>set fwd csum
testpmd>start
-16. Rerun step 11-14.
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,128,256,512
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+
+Test Case 7: Loopback split ring vectorized path multi-queues payload check with server mode and dsa dpdk driver
+----------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split ring
+vectorized path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
+
+
+1. bind 3 dsa port to vfio-pci like common step 1::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u e7:01.0 ec:01.0 f1:01.0
+ ./usertools/dpdk-devbind.py -b vfio-pci e7:01.0 ec:01.0 f1:01.0
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=8 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-17. Quit and relaunch virtio with split ring vectorized path as below::
+3. Launch virtio-user with split ring vectorized path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=0,vectorized=1,server=1 \
@@ -274,30 +366,28 @@ Both iova as VA and PA mode test.
testpmd>set fwd csum
testpmd>start
-18. Rerun step 11-14.
-
-19. Quit and relaunch vhost with diff channel::
-
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0 -a 0000:ec:01.0 -a 0000:f1:01.0 \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0;txq1;txq2;txq3;txq4;txq5;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore11@0000:e7:01.0-q0,lcore12@0000:e7:01.0-q0,lcore13@0000:ec:01.0-q1,lcore13@0000:f1:01.0-q2,lcore14@0000:ec:01.0-q1,lcore14@0000:f1:01.0-q2,lcore15@0000:ec:01.0-q1,lcore15@0000:f1:01.0-q2]
+4. Attach pdump secondary process to primary process by same file-prefix::
-20. Rerun steps 11-14.
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
-21. Quit and relaunch vhost w/ iova=pa::
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0 \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0;txq1;txq2;txq3;txq4;txq5;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=pa -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore11@0000:e7:01.0-q0,lcore12@0000:e7:01.0-q0,lcore13@0000:e7:01.0-q1,lcore13@0000:e7:01.0-q2,lcore14@0000:e7:01.0-q1,lcore14@0000:e7:01.0-q2,lcore15@0000:e7:01.0-q1,lcore15@0000:e7:01.0-q2]
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,128,256,512
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
-22. Rerun steps 11-14.
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
-Test Case 4: Loopback packed ring all path server mode and multi-queues payload check with dsa dpdk driver
-------------------------------------------------------------------------------------------------------------
+Test Case 8: Loopback packed ring inorder mergeable path multi-queues payload check with server mode and dsa dpdk driver
+------------------------------------------------------------------------------------------------------------------------
This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
-all path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver. Both iova as VA and PA mode test.
+inorder mergeable path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
+
1. bind 2 dsa port to vfio-pci like common step 1::
@@ -308,11 +398,10 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
2. Launch vhost::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=4 \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0;txq1;txq2;txq3;txq4;txq5;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore11@0000:e7:01.0-q0,lcore12@0000:e7:01.0-q1,lcore13@0000:e7:01.0-q2,lcore14@0000:e7:01.0-q3]
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-3. Launch virtio-user with packed ring mergeable inorder path::
+3. Launch virtio-user with packed ring inorder mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=1,packed_vq=1,server=1 \
@@ -337,9 +426,25 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
6. Quit pdump, check all the packets length are 6192 Byte in the pcap file and the payload in receive packets are same.
-7. Quit and relaunch vhost and rerun step 4-6.
+Test Case 9: Loopback packed ring mergeable path multi-queues payload check with server mode and dsa dpdk driver
+----------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
+mergeable path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
+
+
+1. bind 2 dsa port to vfio-pci like common step 1::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u e7:01.0 ec:01.0
+ ./usertools/dpdk-devbind.py -b vfio-pci e7:01.0 ec:01.0
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=4 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-8. Quit and relaunch virtio with packed ring mergeable path as below::
+3. Launch virtio-user with packed ring mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=0,packed_vq=1,server=1 \
@@ -347,9 +452,42 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
testpmd>set fwd csum
testpmd>start
-9. Rerun steps 4-7.
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,64,64,2000,2000,2000
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 6192 Byte in the pcap file and the payload in receive packets are same.
-10. Quit and relaunch virtio with packed ring non-mergeable path as below::
+Test Case 10: Loopback packed ring non-mergeable path multi-queues payload check with server mode and dsa dpdk driver
+---------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
+non-mergeable path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
+
+
+1. bind 2 dsa port to vfio-pci like common step 1::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u e7:01.0 ec:01.0
+ ./usertools/dpdk-devbind.py -b vfio-pci e7:01.0 ec:01.0
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=4 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+3. Launch virtio-user with packed ring non-mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=0,packed_vq=1,server=1 \
@@ -357,9 +495,13 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
testpmd>set fwd csum
testpmd>start
-11. Rerun step 4.
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
-12. Send pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
testpmd> set fwd csum
testpmd> set txpkts 64,128,256,512
@@ -368,11 +510,27 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
testpmd> show port stats all
testpmd> stop
-13. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+
+Test Case 11: Loopback packed ring inorder non-mergeable path multi-queues payload check with server mode and dsa dpdk driver
+-----------------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
+inorder non-mergeable path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
+
+
+1. bind 2 dsa port to vfio-pci like common step 1::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u e7:01.0 ec:01.0
+ ./usertools/dpdk-devbind.py -b vfio-pci e7:01.0 ec:01.0
+
+2. Launch vhost::
-14. Quit and relaunch vhost and rerun step 11-13.
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=4 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-15. Quit and relaunch virtio with packed ring inorder non-mergeable path as below::
+3. Launch virtio-user with packed ring inorder non-mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=1,packed_vq=1,server=1 \
@@ -380,9 +538,42 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
testpmd>set fwd csum
testpmd>start
-16. Rerun step 11-14.
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,128,256,512
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+
+Test Case 12: Loopback packed ring vectorized path multi-queues payload check with server mode and dsa dpdk driver
+------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
+vectorized path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa dpdk driver.
+
+
+1. bind 2 dsa port to vfio-pci like common step 1::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u e7:01.0 ec:01.0
+ ./usertools/dpdk-devbind.py -b vfio-pci e7:01.0 ec:01.0
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=4 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-17. Quit and relaunch virtio with packed ring vectorized path as below::
+3. Launch virtio-user with packed ring vectorized path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci --force-max-simd-bitwidth=512 \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=1,packed_vq=1,vectorized=1,server=1 \
@@ -390,9 +581,42 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
testpmd>set fwd csum
testpmd>start
-18. Rerun step 11-14.
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,128,256,512
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+
+Test Case 13: Loopback packed ring vectorized path and ring size is not power of 2 multi-queues payload check with server mode and dsa dpdk driver
+--------------------------------------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
+vectorized path multi-queues with server mode and ring size is not power of 2 when vhost uses the asynchronous operations with dsa dpdk driver.
+
+
+1. bind 2 dsa port to vfio-pci like common step 1::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u e7:01.0 ec:01.0
+ ./usertools/dpdk-devbind.py -b vfio-pci e7:01.0 ec:01.0
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=4 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@0000:e7:01.0-q0;txq1@0000:e7:01.0-q0;txq2@0000:e7:01.0-q0;txq3@0000:e7:01.0-q0;txq4@0000:e7:01.0-q1;txq5@0000:e7:01.0-q1;rxq2@0000:e7:01.0-q2;rxq3@0000:e7:01.0-q2;rxq4@0000:e7:01.0-q3;rxq5@0000:e7:01.0-q3;rxq6@0000:e7:01.0-q3;rxq7@0000:e7:01.0-q3]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-19. Quit and relaunch virtio with packed ring vectorized path and ring size is not power of 2 as below::
+3. Launch virtio-user with packed ring vectorized path and ring size is not power of 2::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci --force-max-simd-bitwidth=512 \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=1,packed_vq=1,vectorized=1,queue_size=1025,server=1 \
@@ -400,28 +624,25 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
testpmd>set fwd csum
testpmd>start
-20. Rerun step 11-14.
-
-21. Quit and relaunch vhost with diff channel::
-
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=2 -a 0000:ec:01.0,max_queues=2 \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0;txq1;txq2;txq3;txq4;txq5;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=va -- -i --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore11@0000:e7:01.0-q0,lcore11@0000:ec:01.0-q1]
+4. Attach pdump secondary process to primary process by same file-prefix::
-22. Rerun steps 11-14.
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
-23. Quit and relaunch vhost w/ iova=pa::
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 -a 0000:e7:01.0,max_queues=4 \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0;txq1;txq2;txq3;txq4;txq5;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=pa -- -i --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore11@0000:e7:01.0-q1,lcore11@0000:e7:01.0-q3]
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,128,256,512
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
-24. Rerun steps 11-14.
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
-Test Case 5: Loopback split ring server mode large chain packets stress test with dsa kernel driver
----------------------------------------------------------------------------------------------------
+Test Case 14: Loopback split ring server mode large chain packets stress test with dsa kernel driver
+----------------------------------------------------------------------------------------------------
This is a stress test case about forwading large chain packets in loopback vhost-user/virtio-user split ring with server mode
when vhost uses the asynchronous enqueue and dequeue operations with dsa kernel driver.
@@ -430,14 +651,14 @@ when vhost uses the asynchronous enqueue and dequeue operations with dsa kernel
ls /dev/dsa #check wq configure, reset if exist
./usertools/dpdk-devbind.py -u 6a:01.0
./usertools/dpdk-devbind.py -b idxd 6a:01.0
- <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 4 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 1 0
ls /dev/dsa #check wq configure success
2. Launch vhost::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -n 4 -l 2-3 --file-prefix=vhost --no-pci \
- --vdev 'net_vhost0,iface=vhost-net0,queues=1,dmas=[txq0;rxq0],client=1' \
- --iova=va -- -i --nb-cores=1 --txd=1024 --rxd=1024 --mbuf-size=45535 --lcore-dma=[lcore3@wq0.2]
+ --vdev 'net_vhost0,iface=vhost-net0,queues=1,dmas=[txq0@wq0.0;rxq0@wq0.0],client=1' \
+ --iova=va -- -i --nb-cores=1 --txd=1024 --rxd=1024 --mbuf-size=65535
3. launch virtio and start testpmd::
@@ -448,11 +669,11 @@ when vhost uses the asynchronous enqueue and dequeue operations with dsa kernel
4. Send large pkts from vhost::
- testpmd>set txpkts 45535,45535,45535,45535,45535
+ testpmd>set txpkts 65535,65535
testpmd>start tx_first 32
testpmd>show port stats all
-Test Case 6: Loopback packed ring server mode large chain packets stress test with dsa kernel driver
+Test Case 15: Loopback packed ring server mode large chain packets stress test with dsa kernel driver
-----------------------------------------------------------------------------------------------------
This is a stress test case about forwading large chain packets in loopback vhost-user/virtio-user packed ring with server mode
when vhost uses the asynchronous operations with dsa kernel driver.
@@ -468,8 +689,8 @@ when vhost uses the asynchronous operations with dsa kernel driver.
2. Launch vhost::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-3 -n 4 --no-pci \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=1,dmas=[txq0;rxq0],client=1' \
- --iova=va -- -i --nb-cores=1 --mbuf-size=45535 --lcore-dma=[lcore3@wq0.0]
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=1,dmas=[txq0@wq0.0;rxq0@wq0.0],client=1' \
+ --iova=va -- -i --nb-cores=1 --mbuf-size=65535
3. launch virtio and start testpmd::
@@ -480,14 +701,14 @@ when vhost uses the asynchronous operations with dsa kernel driver.
4. Send large pkts from vhost and check the stats::
- testpmd>set txpkts 45535,45535,45535,45535,45535
+ testpmd>set txpkts 65535,65535
testpmd>start tx_first 32
testpmd>show port stats all
-Test Case 7: Loopback split ring all path server mode and multi-queues payload check with dsa kernel driver
--------------------------------------------------------------------------------------------------------------
+Test Case 16: Loopback split ring inorder mergeable path multi-queues payload check with server mode and dsa kernel driver
+--------------------------------------------------------------------------------------------------------------------------
This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split ring
-all path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa kernel driver.
+inorder mergeable path multi-queues with server mode when vhost uses the asynchronous operations with dsa kernel driver.
1. bind 2 dsa port to idxd like common step 2::
@@ -501,11 +722,10 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
2. Launch vhost::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 --no-pci \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0;txq1;txq2;txq3;txq4;txq5;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore11@wq0.0,lcore12@wq0.0,lcore13@wq0.1,lcore13@wq0.2,lcore14@wq0.1,lcore14@wq0.2,lcore15@wq0.1,lcore15@wq0.2]
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-3. Launch virtio-user with split ring mergeable inorder path::
+3. Launch virtio-user with split ring inorder mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 5-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=1,server=1 \
@@ -530,29 +750,102 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
6. Quit pdump, check all the packets length are 6192 Byte in the pcap file and the payload in receive packets are same.
-7. Quit and relaunch vhost and rerun step 4-6.
+7. Quit and relaunch vhost w/ diff channel::
-8. Quit and relaunch virtio with split ring mergeable path as below::
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 5-6 -n 4 --file-prefix=virtio-user0 --no-pci \
- --vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=0,server=1 \
- -- -i --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
+8. Rerun step 4-6.
-9. Rerun steps 4-7.
+Test Case 17: Loopback split ring mergeable path multi-queues payload check with server mode and dsa kernel driver
+------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split ring
+mergeable path multi-queues with server mode when vhost uses the asynchronous operations with dsa kernel driver.
+
+1. bind 2 dsa port to idxd like common step 2::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0
+ ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
+ ls /dev/dsa #check wq configure success
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-10. Quit and relaunch virtio with split ring non-mergeable path as below::
+3. Launch virtio-user with split ring mergeable path::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 5-6 -n 4 --file-prefix=virtio-user0 --no-pci \
+ --vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=0,server=1 \
+ -- -i --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+ testpmd> set fwd csum
+ testpmd> start
+
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,64,64,2000,2000,2000
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 6192 Byte in the pcap file and the payload in receive packets are same.
+
+7. Quit and relaunch vhost w/ diff channel::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+8. Rerun step 4-6.
+
+Test Case 18: Loopback split ring non-mergeable path multi-queues payload check with server mode and dsa kernel driver
+----------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split ring
+non-mergeable path multi-queues with server mode when vhost uses the asynchronous operations with dsa kernel driver.
+
+1. bind 2 dsa port to idxd like common step 2::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0
+ ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
+ ls /dev/dsa #check wq configure success
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+3. Launch virtio-user with split ring non-mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 5-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=0,server=1 \
-- -i --enable-hw-vlan-strip --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
+ testpmd> set fwd csum
+ testpmd> start
-11. Rerun step 4.
+4. Attach pdump secondary process to primary process by same file-prefix::
-12. Send pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
testpmd> set fwd csum
testpmd> set txpkts 64,128,256,512
@@ -561,67 +854,149 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
testpmd> show port stats all
testpmd> stop
-13. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
-14. Quit and relaunch vhost and rerun step 11-13.
+7. Quit and relaunch vhost w/ diff channel::
-15. Quit and relaunch virtio with split ring inorder non-mergeable path as below::
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+8. Rerun step 4-6.
+
+Test Case 19: Loopback split ring inorder non-mergeable path multi-queues payload check with server mode and dsa kernel driver
+------------------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split ring
+inorder non-mergeable path multi-queues with server mode when vhost uses the asynchronous operations with dsa kernel driver.
+
+1. bind 2 dsa port to idxd like common step 2::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0
+ ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
+ ls /dev/dsa #check wq configure success
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+3. Launch virtio-user with split ring inorder non-mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 5-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=1,server=1 \
-- -i --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
+ testpmd> set fwd csum
+ testpmd> start
-16. Rerun step 11-14.
+4. Attach pdump secondary process to primary process by same file-prefix::
-17. Quit and relaunch virtio with split ring vectorized path as below::
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,128,256,512
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+
+7. Quit and relaunch vhost w/ diff channel::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+8. Rerun step 4-6.
+
+Test Case 20: Loopback split ring vectorized path multi-queues payload check with server mode and dsa kernel driver
+-------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split ring
+vectorized path multi-queues with server mode when vhost uses the asynchronous operations with dsa kernel driver.
+
+1. bind 2 dsa port to idxd like common step 2::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0
+ ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
+ ls /dev/dsa #check wq configure success
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+3. Launch virtio-user with split ring vectorized path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 5-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=0,vectorized=1,server=1 \
-- -i --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
+ testpmd> set fwd csum
+ testpmd> start
+
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
-18. Rerun step 11-14.
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,128,256,512
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
-19. Quit and relaunch vhost with diff channel::
+7. Quit and relaunch vhost w/ diff channel::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-18 -n 4 --no-pci \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[rxq0;rxq1;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore11@wq0.0,lcore12@wq0.0,lcore13@wq0.1,lcore13@wq1.0,lcore14@wq0.1,lcore14@wq1.0,lcore15@wq0.1,lcore15@wq1.0]
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=5 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-20. Rerun steps 11-14.
+8. Rerun step 4-6.
-Test Case 8: Loopback packed ring all path server mode and multi-queues payload check with dsa kernel driver
--------------------------------------------------------------------------------------------------------------
+Test Case 21: Loopback packed ring inorder mergeable path multi-queues payload check with server mode and dsa kernel driver
+---------------------------------------------------------------------------------------------------------------------------
This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
-all path multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa kernel driver.
+inorder mergeable path multi-queues with server mode when vhost uses the asynchronous operations with dsa kernel driver.
1. bind 8 dsa port to idxd like common step 2::
ls /dev/dsa #check wq configure, reset if exist
./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0
./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
- <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 4 0
- <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 4 1
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
ls /dev/dsa #check wq configure success
2. Launch vhost::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0;txq1;txq2;txq3;txq4;txq5;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore11@wq0.0,lcore12@wq0.1,lcore13@wq0.2,lcore14@wq0.3]
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-3. Launch virtio-user with packed ring mergeable inorder path::
+3. Launch virtio-user with packed ring inorder mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=1,packed_vq=1,server=1 \
-- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
+ testpmd> set fwd csum
+ testpmd> start
4. Attach pdump secondary process to primary process by same file-prefix::
@@ -640,29 +1015,102 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
6. Quit pdump, check all the packets length are 6192 Byte in the pcap file and the payload in receive packets are same.
-7. Quit and relaunch vhost and rerun step 4-6.
+7. Quit and relaunch vhost w/ diff channel::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-8. Quit and relaunch virtio with packed ring mergeable path as below::
+8.Rerun step 4-6.
+
+Test Case 22: Loopback packed ring mergeable path multi-queues payload check with server mode and dsa kernel driver
+-------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
+mergeable path multi-queues with server mode when vhost uses the asynchronous operations with dsa kernel driver.
+
+1. bind 8 dsa port to idxd like common step 2::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0
+ ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
+ ls /dev/dsa #check wq configure success
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+3. Launch virtio-user with packed ring mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=0,packed_vq=1,server=1 \
-- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
+ testpmd> set fwd csum
+ testpmd> start
+
+4. Attach pdump secondary process to primary process by same file-prefix::
-9. Rerun steps 4-7.
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
-10. Quit and relaunch virtio with packed ring non-mergeable path as below::
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,64,64,2000,2000,2000
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 6192 Byte in the pcap file and the payload in receive packets are same.
+
+7. Quit and relaunch vhost w/ diff channel::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+8.Rerun step 4-6.
+
+Test Case 23: Loopback packed ring non-mergeable path multi-queues payload check with server mode and dsa kernel driver
+-----------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
+non-mergeable path multi-queues with server mode when vhost uses the asynchronous operations with dsa kernel driver.
+
+1. bind 8 dsa port to idxd like common step 2::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0
+ ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
+ ls /dev/dsa #check wq configure success
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+3. Launch virtio-user with packed ring non-mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=0,packed_vq=1,server=1 \
-- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
+ testpmd> set fwd csum
+ testpmd> start
-11. Rerun step 4.
+4. Attach pdump secondary process to primary process by same file-prefix::
-12. Send pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
testpmd> set fwd csum
testpmd> set txpkts 64,128,256,512
@@ -671,59 +1119,183 @@ all path multi-queues with server mode when vhost uses the asynchronous enqueue
testpmd> show port stats all
testpmd> stop
-13. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+
+7. Quit and relaunch vhost w/ diff channel::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+8.Rerun step 4-6.
+
+Test Case 24: Loopback packed ring inorder non-mergeable path multi-queues payload check with server mode and dsa kernel driver
+-------------------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
+inorder non-mergeable path multi-queues with server mode when vhost uses the asynchronous operations with dsa kernel driver.
+
+1. bind 8 dsa port to idxd like common step 2::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0
+ ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
+ ls /dev/dsa #check wq configure success
+
+2. Launch vhost::
-14. Quit and relaunch vhost and rerun step 11-13.
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-15. Quit and relaunch virtio with packed ring inorder non-mergeable path as below::
+3. Launch virtio-user with packed ring inorder non-mergeable path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=1,packed_vq=1,server=1 \
-- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
+ testpmd> set fwd csum
+ testpmd> start
+
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,128,256,512
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+
+7. Quit and relaunch vhost w/ diff channel::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+8.Rerun step 4-6.
+
+Test Case 25: Loopback packed ring vectorized path multi-queues payload check with server mode and dsa kernel driver
+--------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
+vectorized path multi-queues with server mode when vhost uses the asynchronous operations with dsa kernel driver.
+
+1. bind 8 dsa port to idxd like common step 2::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0
+ ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
+ ls /dev/dsa #check wq configure success
-16. Rerun step 11-14.
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-17. Quit and relaunch virtio with packed ring vectorized path as below::
+3. Launch virtio-user with packed ring vectorized path::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci --force-max-simd-bitwidth=512 \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=1,packed_vq=1,vectorized=1,server=1 \
-- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
+ testpmd> set fwd csum
+ testpmd> start
+
+4. Attach pdump secondary process to primary process by same file-prefix::
-18. Rerun step 11-14.
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,128,256,512
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
+
+7. Quit and relaunch vhost w/ diff channel::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+
+8.Rerun step 4-6.
+
+Test Case 26: Loopback packed ring vectorized path and ring size is not power of 2 multi-queues payload check with server mode and dsa kernel driver
+----------------------------------------------------------------------------------------------------------------------------------------------------
+This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user packed ring
+vectorized path multi-queues with server mode and ring size is not power of 2 when vhost uses the asynchronous operations with dsa kernel driver.
+
+1. bind 8 dsa port to idxd like common step 2::
+
+ ls /dev/dsa #check wq configure, reset if exist
+ ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0
+ ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
+ <dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
+ ls /dev/dsa #check wq configure success
+
+2. Launch vhost::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@wq0.1;txq5@wq0.1;rxq2@wq1.0;rxq3@wq1.0;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-19. Quit and relaunch virtio with packed ring vectorized path and ring size is not power of 2 as below::
+3. Launch virtio-user with packed ring vectorized path and ring size is not power of 2::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=virtio-user0 --no-pci --force-max-simd-bitwidth=512 \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=0,in_order=1,packed_vq=1,vectorized=1,queue_size=1025,server=1 \
- -- -i --nb-cores=2 --rxq=8 --txq=8 --txd=1025 --rxd=1025
+ -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1025 --rxd=1025
testpmd>set fwd csum
testpmd>start
-20. Rerun step 11-14.
+4. Attach pdump secondary process to primary process by same file-prefix::
+
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-pdump -v --file-prefix=virtio-user0 -- \
+ --pdump 'device_id=net_virtio_user0,queue=0,rx-dev=./pdump-virtio-rx-0.pcap,mbuf-size=8000' \
+ --pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
+
+5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+
+ testpmd> set fwd csum
+ testpmd> set txpkts 64,128,256,512
+ testpmd> set burst 1
+ testpmd> start tx_first 1
+ testpmd> show port stats all
+ testpmd> stop
+
+6. Quit pdump, check all the packets length are 960 Byte in the pcap file and the payload in receive packets are same.
-21. Quit and relaunch vhost with diff channel::
+7. Quit and relaunch vhost w/ diff channel::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --no-pci \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0;txq1;txq2;txq3;txq4;txq5;txq6;txq7;rxq0;rxq1;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore11@wq0.0,lcore11@wq1.0,lcore12@wq0.1,lcore12@wq1.1,lcore13@wq0.2,lcore13@wq1.2,lcore14@wq0.3,lcore14@wq1.3]
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.1;txq2@wq0.2;txq3@wq0.3;txq4@wq0.4;txq5@wq0.5;rxq2@wq1.2;rxq3@wq1.3;rxq4@wq1.4;rxq5@wq1.5;rxq6@wq1.6;rxq7@wq1.7]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
-22. Rerun steps 3-6.
+8.Rerun step 4-6.
-Test Case 9: Loopback split and packed ring server mode multi-queues and mergeable path payload check with dsa dpdk and kernel driver
---------------------------------------------------------------------------------------------------------------------------------------
-This case tests the payload is valid after forwading large chain packets in loopback vhost-user/virtio-user split and packed ring
-multi-queues with server mode when vhost uses the asynchronous enqueue and dequeue operations with dsa kernel driver.
+Test Case 27: PV split and packed ring server mode test txonly mode with dsa dpdk and kernel driver
+---------------------------------------------------------------------------------------------------
1. bind 2 dsa device to idxd and 2 dsa device to vfio-pci like common step 1-2::
ls /dev/dsa #check wq configure, reset if exist
./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0 f1:01.0 f6:01.0
- ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0 f1:01.0 f6:01.0
+ ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0
./usertools/dpdk-devbind.py -b vfio-pci f1:01.0 f6:01.0
<dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 0
<dpdk dir># ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 8 1
@@ -731,17 +1303,16 @@ multi-queues with server mode when vhost uses the asynchronous enqueue and deque
2. Launch vhost::
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-3 -n 4 --file-prefix=vhost -a 0000:f1:01.0,max_queues=4 -a 0000:f6:01.0,max_queues=4 \
- --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0;txq1;txq2;txq3;txq4;txq5;txq6;txq7;rxq0;rxq1;rxq2;rxq3;rxq4;rxq5;rxq6;rxq7]' \
- --iova=va -- -i --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024 \
- --lcore-dma=[lcore3@wq0.0,lcore3@wq0.1,lcore3@wq1.0,lcore3@wq1.1,lcore3@0000:f1:01.0-q0,lcore3@0000:f1:01.0-q2,lcore3@0000:f6:01.0-q3]
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 2-6 -n 4 --file-prefix=vhost -a 0000:f1:01.0,max_queues=2 \
+ --vdev 'eth_vhost0,iface=vhost-net0,queues=8,client=1,dmas=[txq0@wq0.0;txq1@wq0.0;txq2@wq0.0;txq3@wq0.0;txq4@0000:f1:01.0-q0;txq5@0000:f1:01.0-q0;rxq2@0000:f1:01.0-q1;rxq3@0000:f1:01.0-q1;rxq4@wq1.1;rxq5@wq1.1;rxq6@wq1.1;rxq7@wq1.1]' \
+ --iova=va -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
3. Launch virtio-user with split ring mergeable inorder path::
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 4-5 -n 4 --file-prefix=virtio-user --no-pci \
+ <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 10-14 -n 4 --file-prefix=virtio-user --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=1,server=1 \
- -- -i --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
+ -- -i --nb-cores=4 --rxq=8 --txq=8 --txd=1024 --rxd=1024
+ testpmd>set fwd rxonly
testpmd>start
4. Attach pdump secondary process to primary process by same file-prefix::
@@ -750,30 +1321,20 @@ multi-queues with server mode when vhost uses the asynchronous enqueue and deque
--pdump 'device_id=net_virtio_user0,queue=0,rx-dev=/tmp/pdump-virtio-rx-0.pcap,mbuf-size=8000'
--pdump 'device_id=net_virtio_user0,queue=3,rx-dev=./pdump-virtio-rx-3.pcap,mbuf-size=8000'
-5. Send large pkts from vhost, check loopback performance can get expected and each queue can receive packets::
+5. Send large pkts from vhost::
- testpmd>set fwd csum
- testpmd>set txpkts 64,64,64,2000,2000,2000
- testpmd>set burst 1
- testpmd>start tx_first 1
- testpmd>show port stats all
- testpmd>stop
+ testpmd> set fwd txonly
+ testpmd> async_vhost tx poll completed on
+ testpmd>set txpkts 64,64,64,2000,2000,2000
+ testpmd>set burst 1
+ testpmd>start tx_first 1
+ testpmd>stop
-6. Quit pdump and chcek all the packets length is 6192 and the payload of all packets are same in the pcap file.
+6. Quit pdump, check all the packets length are 6192 Byte in the pcap file.
7. Quit and relaunch vhost and rerun step 4-6.
-8. Quit and relaunch virtio with split ring mergeable path as below::
-
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 4-5 -n 4 --file-prefix=virtio-user --no-pci \
- --vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=0,server=1 \
- -- -i --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
-
-9. Stop vhost and rerun step 4-7.
-
-10. Quit and relaunch virtio with packed ring mergeable inorder path as below::
+8. Quit and relaunch virtio with packed ring mergeable inorder path as below::
<dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 4-5 -n 4 --file-prefix=virtio-user --no-pci \
--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=1,packed_vq=1,server=1 \
@@ -781,14 +1342,4 @@ multi-queues with server mode when vhost uses the asynchronous enqueue and deque
testpmd>set fwd csum
testpmd>start
-11. Stop vhost and rerun step 4-7.
-
-12. Quit and relaunch virtio with packed ring mergeable path as below::
-
- <dpdk dir># ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 4-5 -n 4 --file-prefix=virtio-user --no-pci \
- --vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=8,mrg_rxbuf=1,in_order=0,packed_vq=1,server=1 \
- -- -i --nb-cores=1 --rxq=8 --txq=8 --txd=1024 --rxd=1024
- testpmd>set fwd csum
- testpmd>start
-
-13. Stop vhost and rerun step 4-7.
+9. Stop vhost and rerun step 4-7.