From mboxrd@z Thu Jan 1 00:00:00 1970
Return-Path: <bugzilla@dpdk.org>
Received: by dpdk.org (Postfix, from userid 33)
id 327C11B437; Mon, 14 Jan 2019 10:27:35 +0100 (CET)
From: bugzilla@dpdk.org
To: dev@dpdk.org
Date: Mon, 14 Jan 2019 09:27:34 +0000
X-Bugzilla-Reason: AssignedTo
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: DPDK
X-Bugzilla-Component: testpmd
X-Bugzilla-Version: 19.02
X-Bugzilla-Keywords:
X-Bugzilla-Severity: critical
X-Bugzilla-Who: alialnu@mellanox.com
X-Bugzilla-Status: CONFIRMED
X-Bugzilla-Resolution:
X-Bugzilla-Priority: Normal
X-Bugzilla-Assigned-To: dev@dpdk.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: bug_id short_desc product version rep_platform
op_sys bug_status bug_severity priority component assigned_to reporter
target_milestone
Message-ID: <bug-188-3@http.bugs.dpdk.org/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Bugzilla-URL: http://bugs.dpdk.org/
Auto-Submitted: auto-generated
X-Auto-Response-Suppress: All
MIME-Version: 1.0
Subject: [dpdk-dev] [Bug 188] significant multi-core performance degradation
with SR-IOV/macswap in some cases
X-BeenThere: dev@dpdk.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <https://mails.dpdk.org/options/dev>,
<mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://mails.dpdk.org/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <https://mails.dpdk.org/listinfo/dev>,
<mailto:dev-request@dpdk.org?subject=subscribe>
X-List-Received-Date: Mon, 14 Jan 2019 09:27:35 -0000
https://bugs.dpdk.org/show_bug.cgi?id=3D188
Bug ID: 188
Summary: significant multi-core performance degradation with
SR-IOV/macswap in some cases
Product: DPDK
Version: 19.02
Hardware: x86
OS: Linux
Status: CONFIRMED
Severity: critical
Priority: Normal
Component: testpmd
Assignee: dev@dpdk.org
Reporter: alialnu@mellanox.com
Target Milestone: ---
Hi,
I'm seeing a performance degradation with testpmd running with multiple cor=
es.
I saw the issue with ConnectX-3, ConnectX-4 Lx, and ConnectX-5.
### For ConnectX-3:
- With 2 Cores, I see a degradation from 19.2 mpps to 13.5 mpps.
testpmd command:
`MLX4_INLINE_RECV_SIZE=3D64 ./build/app/test-pmd/testpmd -c 0x7 -n 4 -w
0002:00:02.0 --vdev=3Dnet_vdev_netvsc,ignore=3D1 --socket-mem=3D2048 --
--port-numa-config=3D0,0 --socket-num=3D0 --burst=3D64 --txd=3D256 --rxd=3D=
256
--mbcache=3D512 --rxq=3D2 --txq=3D2 --nb-cores=3D2 -i -a --forward-mode=3D=
macswap`
- With 4 Cores, I see a degradation from 19.2 mpps - 13.9 mpps.
testpmd command:
`MLX4_INLINE_RECV_SIZE=3D64 ./build/app/test-pmd/testpmd -c 0x1f -n 4 -w
0002:00:02.0 --vdev=3Dnet_vdev_netvsc,ignore=3D1 --socket-mem=3D2048 --
--port-numa-config=3D0,0 --socket-num=3D0 --burst=3D64 --txd=3D256 --rxd=3D=
256
--mbcache=3D512 --rxq=3D2 --txq=3D2 --nb-cores=3D4 -i -a --forward-mode=3D=
macswap`
### For ConnectX-4 Lx:
- With 2 Cores, I see a degradation from 32.9 mpps to 26.8 mpps.
testpmd command:
`./build/app/test-pmd/testpmd -c 0x7 -n 4 -w 0003:00:02.0
--vdev=3D"net_vdev_netvsc,ignore=3D1" -- --burst=3D64 --txd=3D1024 --rxd=3D=
256
--mbcache=3D512 --rxq=3D2 --txq=3D2 --nb-cores=3D2 --rss-udp --port-topolog=
y=3Dchained
--forward-mode=3Dmacswap -i -a`
- With 4 Cores, I see a degradation from 34.8 mpps to 19 mpps.
testpmd command:
`./build/app/test-pmd/testpmd -c 0x1f -n 4 -w 0003:00:02.0
--vdev=3D"net_vdev_netvsc,ignore=3D1" -- --burst=3D64 --txd=3D8192 --rxd=3D=
256
--mbcache=3D512 --rxq=3D4 --txq=3D4 --nb-cores=3D4 --rss-udp --port-topolog=
y=3Dchained
--forward-mode=3Dmacswap -i -a`
Server specs:
OS: Ubuntu 16.04.2 LTS
Kernel: 4.15.0-rc3
I'm using IXIA as a traffic generator, and to get the forwarding performanc=
e.
DPDK version: master:a89fbda
The cause of the degradation is the patch:
```
commit 62b52877adbe8d3ad19615b0827e7725e51feadc
Author: Qi Zhang <qi.z.zhang@intel.com>
Date: Sun Dec 16 08:58:35 2018 +0800
app/testpmd: batch MAC swap for performance on x86
Do four packets macswap in same loop iterate to squeeze more
CPU cycles.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
```
--=20
You are receiving this mail because:
You are the assignee for the bug.=