test suite reviews and discussions
 help / color / mirror / Atom feed
From: Yinan <yinan.wang@intel.com>
To: dts@dpdk.org
Cc: Wang Yinan <yinan.wang@intel.com>
Subject: [dts] [PATCH 10/11 v1] test_plans: add test cases for virtio_event_idx_interrupt
Date: Fri, 28 Feb 2020 06:09:46 +0000	[thread overview]
Message-ID: <20200228060947.26001-11-yinan.wang@intel.com> (raw)
In-Reply-To: <20200228060947.26001-1-yinan.wang@intel.com>

From: Wang Yinan <yinan.wang@intel.com>

Signed-off-by: Wang Yinan <yinan.wang@intel.com>
---
 .../virtio_event_idx_interrupt_test_plan.rst  | 123 ++++++++++++++++--
 1 file changed, 115 insertions(+), 8 deletions(-)

diff --git a/test_plans/virtio_event_idx_interrupt_test_plan.rst b/test_plans/virtio_event_idx_interrupt_test_plan.rst
index e51d808..8293032 100644
--- a/test_plans/virtio_event_idx_interrupt_test_plan.rst
+++ b/test_plans/virtio_event_idx_interrupt_test_plan.rst
@@ -37,16 +37,17 @@ virtio event idx interrupt mode test plan
 Description
 ===========
 
-This feature is to suppress interrupts for performance improvement, need compare interrupt times with and without
-virtio event idx enabled. Also need cover driver reload and live migration test.
+This feature is to suppress interrupts for performance improvement, need compare
+interrupt times with and without virtio event idx enabled. Also need cover driver
+reload test. For packed virtqueue test, need using qemu version > 4.2.0.
 
 Test flow
 =========
 
 TG --> NIC --> Vhost-user --> Virtio-net
 
-Test Case 1: compare interrupt times with and without virtio event idx enabled
-==============================================================================
+Test Case 1: Compare interrupt times with and without split ring virtio event idx enabled
+=========================================================================================
 
 1. Bind one nic port to igb_uio, then launch the vhost sample by below commands::
 
@@ -78,8 +79,8 @@ Test Case 1: compare interrupt times with and without virtio event idx enabled
 
 6. Compare interrupt times between virtio event_idx enabled and virtio event_idx disabled.
 
-Test Case 2: virtio-pci driver reload test
-==========================================
+Test Case 2: Split ring virtio-pci driver reload test
+=====================================================
 
 1. Bind one nic port to igb_uio, then launch the vhost sample by below commands::
 
@@ -116,8 +117,8 @@ Test Case 2: virtio-pci driver reload test
 
 6. Rerun step4 and step5 100 times to check event idx workable after driver reload.
 
-Test Case 3: wake up virtio-net cores with event idx interrupt mode 16 queues test
-==================================================================================
+Test Case 3: Wake up split ring virtio-net cores with event idx interrupt mode 16 queues test
+=============================================================================================
 
 1. Bind one nic port to igb_uio, then launch the vhost sample by below commands::
 
@@ -150,3 +151,109 @@ Test Case 3: wake up virtio-net cores with event idx interrupt mode 16 queues te
     testpmd>stop
     testpmd>start
     testpmd>stop
+
+Test Case 4: Compare interrupt times with and without packed ring virtio event idx enabled
+==========================================================================================
+
+1. Bind one nic port to igb_uio, then launch the vhost sample by below commands::
+
+    rm -rf vhost-net*
+    ./testpmd -c 0xF0000000 -n 4 --socket-mem 2048,2048 --legacy-mem --file-prefix=vhost --vdev 'net_vhost,iface=vhost-net,queues=1' -- -i
+    --nb-cores=1 --txd=1024 --rxd=1024
+    testpmd>start
+
+2. Launch VM::
+
+    taskset -c 32-33 \
+    qemu-system-x86_64 -name us-vhost-vm1 \
+     -cpu host -enable-kvm -m 2048 -object memory-backend-file,id=mem,size=2048M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc \
+     -smp cores=2,sockets=1 -drive file=/home/osimg/ubuntu16.img  \
+     -monitor unix:/tmp/vm2_monitor.sock,server,nowait -net nic,vlan=2,macaddr=00:00:00:08:e8:aa,addr=1f -net user,vlan=2,hostfwd=tcp:127.0.0.1:6004-:22 \
+     -chardev socket,id=char0,path=./vhost-net -netdev type=vhost-user,id=mynet1,chardev=char0,vhostforce \
+     -device virtio-net-pci,mac=52:54:00:00:00:01,netdev=mynet1,mrg_rxbuf=on,csum=on,gso=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on,packed=on \
+     -vnc :12 -daemonize
+
+3. On VM1, set virtio device IP::
+
+    ifconfig [ens3] 1.1.1.2  # [ens3] is the name of virtio-net
+
+4. Send 10M packets from packet generator to nic, check virtio-net interrupt times by below cmd in VM::
+
+    cat /proc/interrupts
+
+5. Disable virtio event idx feature and rerun step1 ~ step4.
+
+6. Compare interrupt times between virtio event_idx enabled and virtio event_idx disabled.
+
+Test Case 5: Packed ring virtio-pci driver reload test
+======================================================
+
+1. Bind one nic port to igb_uio, then launch the vhost sample by below commands::
+
+    rm -rf vhost-net*
+    ./testpmd -c 0xF0000000 -n 4 --socket-mem 2048,2048 --legacy-mem --file-prefix=vhost --vdev 'net_vhost,iface=vhost-net,queues=1' -- -i --nb-cores=1 --txd=1024 --rxd=1024
+    testpmd>start
+
+2. Launch VM::
+
+    taskset -c 32-33 \
+    qemu-system-x86_64 -name us-vhost-vm1 \
+     -cpu host -enable-kvm -m 2048 -object memory-backend-file,id=mem,size=2048M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc \
+     -smp cores=2,sockets=1 -drive file=/home/osimg/ubuntu16.img  \
+     -monitor unix:/tmp/vm2_monitor.sock,server,nowait -net nic,vlan=2,macaddr=00:00:00:08:e8:aa,addr=1f -net user,vlan=2,hostfwd=tcp:127.0.0.1:6004-:22 \
+     -chardev socket,id=char0,path=./vhost-net -netdev type=vhost-user,id=mynet1,chardev=char0,vhostforce \
+     -device virtio-net-pci,mac=52:54:00:00:00:01,netdev=mynet1,mrg_rxbuf=on,csum=on,gso=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on,packed=on \
+     -vnc :12 -daemonize
+
+3. On VM1, set virtio device IP, send 10M packets from packet generator to nic then check virtio device can receive packets::
+
+    ifconfig [ens3] 1.1.1.2      # [ens3] is the name of virtio-net
+    tcpdump -i [ens3]
+
+4. Reload virtio-net driver by below cmds::
+
+    ifconfig [ens3] down
+    ./dpdk-devbind.py -u [00:03.0]   # [00:03.0] is the pci addr of virtio-net
+    ./dpdk-devbind.py -b virtio-pci [00:03.0]
+
+5. Check virtio device can receive packets again::
+
+    ifconfig [ens3] 1.1.1.2
+    tcpdump -i [ens3]
+
+6. Rerun step4 and step5 100 times to check event idx workable after driver reload.
+
+Test Case 6: Wake up packed ring virtio-net cores with event idx interrupt mode 16 queues test
+==============================================================================================
+
+1. Bind one nic port to igb_uio, then launch the vhost sample by below commands::
+
+    rm -rf vhost-net*
+    ./testpmd -l 1-17 -n 4 --socket-mem 2048,2048 --legacy-mem --file-prefix=vhost --vdev 'net_vhost,iface=vhost-net,queues=16' -- -i --nb-cores=16 --txd=1024 --rxd=1024 --rxq=16 --txq=16
+    testpmd>start
+
+2. Launch VM::
+
+    taskset -c 32-33 \
+    qemu-system-x86_64 -name us-vhost-vm1 \
+     -cpu host -enable-kvm -m 2048 -object memory-backend-file,id=mem,size=2048M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc \
+     -smp cores=16,sockets=1 -drive file=/home/osimg/ubuntu16.img  \
+     -monitor unix:/tmp/vm2_monitor.sock,server,nowait -net nic,vlan=2,macaddr=00:00:00:08:e8:aa,addr=1f -net user,vlan=2,hostfwd=tcp:127.0.0.1:6004-:22 \
+     -chardev socket,id=char0,path=./vhost-net -netdev type=vhost-user,id=mynet1,chardev=char0,vhostforce,queues=16 \
+     -device virtio-net-pci,mac=52:54:00:00:00:01,netdev=mynet1,mrg_rxbuf=on,mq=on,vectors=40,csum=on,gso=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on,packed=on \
+     -vnc :12 -daemonize
+
+3. On VM1, give virtio device ip addr and enable vitio-net with 16 quques::
+
+    ifconfig [ens3] 1.1.1.2           # [ens3] is the name of virtio-net
+    ethtool -L [ens3] combined 16
+
+4. Send 10M different ip addr packets from packet generator to nic, check virtio-net interrupt times by below cmd in VM::
+
+    cat /proc/interrupts
+
+5. After two hours stress test, stop and restart testpmd, check each queue has new packets coming::
+
+    testpmd>stop
+    testpmd>start
+    testpmd>stop
\ No newline at end of file
-- 
2.17.1


  parent reply	other threads:[~2020-02-28 13:15 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-28  6:09 [dts] [PATCH 00/11 v1] test_plans: add packed ring cases Yinan
2020-02-28  6:09 ` [dts] [PATCH 01/11 v1] test_plans: add packed ring cases for loopback_multi_paths_port_restart Yinan
2020-02-28  6:09 ` [dts] [PATCH 02/11 v1] test_plans: add packed ring cases for loopback_multi_queues Yinan
2020-02-28  6:09 ` [dts] [PATCH 03/11 v1] test_plans: add packed ring test case for pvp_virtio_user_2M_hugepages Yinan
2020-02-28  6:09 ` [dts] [PATCH 04/11 v1] test_plans: add packed ring cases for pvp_virtio_user_4k_pages Yinan
2020-02-28  6:09 ` [dts] [PATCH 05/11 v1] test_plans: add packed ring test case for vhost_enqueue_interrupt Yinan
2020-02-28  6:09 ` [dts] [PATCH 06/11 v1] test_plans: add packed ring test cases for vhost_event_idx_interrupt Yinan
2020-02-28  6:09 ` [dts] [PATCH 07/11 v1] test_plans: add packed ring test cases for vhost_user_live_migration Yinan
2020-02-28  6:09 ` [dts] [PATCH 08/11 v1] test_plans: add packed ring test cases for vhost_virtio_pmd_interrupt Yinan
2020-02-28  6:09 ` [dts] [PATCH 09/11 v1] test_plans: add packed ring test cases for vhost_virtio_user_interrupt Yinan
2020-02-28  6:09 ` Yinan [this message]
2020-02-28  6:09 ` [dts] [PATCH 11/11 v1] test_plans: add packed ring test cases for virtio_pvp_regression Yinan
2020-03-03  7:28 ` [dts] [PATCH 00/11 v1] test_plans: add packed ring cases Tu, Lijuan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200228060947.26001-11-yinan.wang@intel.com \
    --to=yinan.wang@intel.com \
    --cc=dts@dpdk.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).