[dpdk-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service

[dpdk-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service

[Alex Yeh (ayeh)]

Hi,
               We are seeing a ovs-vswitchd service crash with segfault in the librte_vhost library when a DPDK application within a guest VM is stopped.

               We are using OVS 2.11.1 on CentOS 7.6 (3.10.0-1062 Linux kernel) with DPDK 18.11.2.

               We are using OVS-DPDK on the host and the guest VM is running a DPDK application. With some traffic, if the application service within the VM is restarted, then OVS crashes.

               This crash is not seen if the guest VM is restarted (instead of stopping the application within the VM).

               The crash trackback (attached below) points to the rte_memcpy_generic() function in rte_memcpy.h. It looks like the crash occurs when vhost is trying to dequeue the packets from the guest VM (as the application in the guest VM has stopped and the huge pages are returned to the guest kernel).

               We have tried enabling iommu in ovs by setting
"other_config:vhost-iommu-support=true" and enabling iommu in qemu using the following configuration in the guest domain XML:
<iommu model='intel'>
    <driver intremap='on'/>
</iommu>
               With iommu enabled ovs-vswitchd still crashes when guest VM restarts the network service.

               Is this a known problem? Anyone else seen a crash like this?  How can we protect the ovs-vswitchd from crashing when a guest VM restarts the network application or service?

Thanks
Alex
------------------------------------------------------------------------

Log:
Oct 7 19:54:16 Branch81-Bravo kernel: [2245909.596635] pmd16[25721]: segfault at 7f4d1d733000 ip 00007f4d2ae5d066 sp 00007f4d1ce65618 error 4 in librte_vhost.so.4[7f4d2ae52000+1a000]
Oct 7 19:54:19 Branch81-Bravo systemd[1]: ovs-vswitchd.service: main process exited, code=killed, status=11/SEGV

Environment:
CentOs 7.6.1810
openvswitch-2.11.1-1.el7.centos.x86_64
openvswitch-kmod-2.11.1-1.el7.centos.x86_64
dpdk-18.11-2.el7.centos.x86_64
3.10.0-1062.4.1.el7.x86_64
qemu-kvm-ev-2.12.0-18.el7.centos_6.1.1

Core dump trace:
(gdb) bt
#-1 0x00007ffff205602e in rte_memcpy_generic (dst=<optimized out>,
src=0x7fffcef3607c, n=<optimized out>)
at /usr/src/debug/dpdk-18.11/x86_64-native-linuxapp-gcc/include/rte_memcpy.h:793
Backtrace stopped: Cannot access memory at address 0x7ffff20558f0

(gdb) list *0x00007ffff205602e
0x7ffff205602e is in rte_memcpy_generic (/usr/src/debug/dpdk-18.11/x86_64-native-linuxapp-gcc/include/rte_memcpy.h:793).
788 }
789
790 /**
791 * For copy with unaligned load
792 */
793 MOVEUNALIGNED_LEFT47(dst, src, n, srcofs);
794
795 /**
796 * Copy whatever left
797 */

(gdb) list *0x00007ffff205c192
0x7ffff205c192 is in rte_vhost_dequeue_burst (/usr/src/debug/dpdk-18.11/lib/librte_vhost/virtio_net.c:1192).
1187 * In zero copy mode, one mbuf can only reference data
1188 * for one or partial of one desc buff.
1189 */
1190 mbuf_avail = cpy_len;
1191 } else {
1192 if (likely(cpy_len > MAX_BATCH_LEN ||
1193 vq->batch_copy_nb_elems >= vq->size ||
1194 (hdr && cur == m))) {
1195 rte_memcpy(rte_pktmbuf_mtod_offset(cur, void *,
1196 mbuf_offset),
(gdb)

Re: [dpdk-dev] [ovs-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service

[Stokes, Ian]

> Hi,
>                We are seeing a ovs-vswitchd service crash with segfault in the
> librte_vhost library when a DPDK application within a guest VM is stopped.
> 
>                We are using OVS 2.11.1 on CentOS 7.6 (3.10.0-1062 Linux kernel) with
> DPDK 18.11.2.

Hi,

Is there a reason you are using OVS 2.11.1 and DPDK 18.11.2?  These are quite old.

As a first step I would recommend using the latest of these branches that have been validated with by the OVS community.

As of now this would be OVS 2.11.4 and DPDK 18.11.9 to check if the issue is still present there my suspicion is that this could be an issue resolved in the DPDK library since 18.11.2.

Regards
Ian

> 
>                We are using OVS-DPDK on the host and the guest VM is running a DPDK
> application. With some traffic, if the application service within the VM is
> restarted, then OVS crashes.
> 
>                This crash is not seen if the guest VM is restarted (instead of stopping
> the application within the VM).
> 
>                The crash trackback (attached below) points to the
> rte_memcpy_generic() function in rte_memcpy.h. It looks like the crash occurs
> when vhost is trying to dequeue the packets from the guest VM (as the
> application in the guest VM has stopped and the huge pages are returned to the
> guest kernel).
> 
>                We have tried enabling iommu in ovs by setting
> "other_config:vhost-iommu-support=true" and enabling iommu in qemu using
> the following configuration in the guest domain XML:
> <iommu model='intel'>
>     <driver intremap='on'/>
> </iommu>
>                With iommu enabled ovs-vswitchd still crashes when guest VM restarts
> the network service.
> 
>                Is this a known problem? Anyone else seen a crash like this?  How can
> we protect the ovs-vswitchd from crashing when a guest VM restarts the
> network application or service?
> 
> Thanks
> Alex
> ------------------------------------------------------------------------
> 
> Log:
> Oct 7 19:54:16 Branch81-Bravo kernel: [2245909.596635] pmd16[25721]:
> segfault at 7f4d1d733000 ip 00007f4d2ae5d066 sp 00007f4d1ce65618 error 4 in
> librte_vhost.so.4[7f4d2ae52000+1a000]
> Oct 7 19:54:19 Branch81-Bravo systemd[1]: ovs-vswitchd.service: main process
> exited, code=killed, status=11/SEGV
> 
> Environment:
> CentOs 7.6.1810
> openvswitch-2.11.1-1.el7.centos.x86_64
> openvswitch-kmod-2.11.1-1.el7.centos.x86_64
> dpdk-18.11-2.el7.centos.x86_64
> 3.10.0-1062.4.1.el7.x86_64
> qemu-kvm-ev-2.12.0-18.el7.centos_6.1.1
> 
> Core dump trace:
> (gdb) bt
> #-1 0x00007ffff205602e in rte_memcpy_generic (dst=<optimized out>,
> src=0x7fffcef3607c, n=<optimized out>)
> at /usr/src/debug/dpdk-18.11/x86_64-native-linuxapp-
> gcc/include/rte_memcpy.h:793
> Backtrace stopped: Cannot access memory at address 0x7ffff20558f0
> 
> (gdb) list *0x00007ffff205602e
> 0x7ffff205602e is in rte_memcpy_generic (/usr/src/debug/dpdk-18.11/x86_64-
> native-linuxapp-gcc/include/rte_memcpy.h:793).
> 788 }
> 789
> 790 /**
> 791 * For copy with unaligned load
> 792 */
> 793 MOVEUNALIGNED_LEFT47(dst, src, n, srcofs);
> 794
> 795 /**
> 796 * Copy whatever left
> 797 */
> 
> (gdb) list *0x00007ffff205c192
> 0x7ffff205c192 is in rte_vhost_dequeue_burst (/usr/src/debug/dpdk-
> 18.11/lib/librte_vhost/virtio_net.c:1192).
> 1187 * In zero copy mode, one mbuf can only reference data
> 1188 * for one or partial of one desc buff.
> 1189 */
> 1190 mbuf_avail = cpy_len;
> 1191 } else {
> 1192 if (likely(cpy_len > MAX_BATCH_LEN ||
> 1193 vq->batch_copy_nb_elems >= vq->size ||
> 1194 (hdr && cur == m))) {
> 1195 rte_memcpy(rte_pktmbuf_mtod_offset(cur, void *,
> 1196 mbuf_offset),
> (gdb)
> 
> _______________________________________________
> dev mailing list
> dev@openvswitch.org
> https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Re: [dpdk-dev] [ovs-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service

[Kevin Traynor]

On 19/11/2020 11:21, Stokes, Ian wrote:
>> Hi,
>>                We are seeing a ovs-vswitchd service crash with segfault in the
>> librte_vhost library when a DPDK application within a guest VM is stopped.
>>
>>                We are using OVS 2.11.1 on CentOS 7.6 (3.10.0-1062 Linux kernel) with
>> DPDK 18.11.2.
> 
> Hi,
> 
> Is there a reason you are using OVS 2.11.1 and DPDK 18.11.2?  These are quite old.
> 
> As a first step I would recommend using the latest of these branches that have been validated with by the OVS community.
> 
> As of now this would be OVS 2.11.4 and DPDK 18.11.9 to check if the issue is still present there my suspicion is that this could be an issue resolved in the DPDK library since 18.11.2.
> 

+1, there's 58 commits in the vhost library on 18.11 branch since
18.11.2, so it might be already fixed. 18.11.10 is the latest release,
while below is in from 18.11.7.

$ git log --oneline v18.11.2..HEAD . | grep crash
90b5ba739f vhost: fix crash on port deletion

If you are planning to continue to use 18.11 for a while, I think you
will want to test the 18.11.11 Release Candidate that will be available
in a few weeks. It is the last planned 18.11 release, so any issues you
find *after* it is released won't be fixed.

Kevin.



> Regards
> Ian
> 
>>
>>                We are using OVS-DPDK on the host and the guest VM is running a DPDK
>> application. With some traffic, if the application service within the VM is
>> restarted, then OVS crashes.
>>
>>                This crash is not seen if the guest VM is restarted (instead of stopping
>> the application within the VM).
>>
>>                The crash trackback (attached below) points to the
>> rte_memcpy_generic() function in rte_memcpy.h. It looks like the crash occurs
>> when vhost is trying to dequeue the packets from the guest VM (as the
>> application in the guest VM has stopped and the huge pages are returned to the
>> guest kernel).
>>
>>                We have tried enabling iommu in ovs by setting
>> "other_config:vhost-iommu-support=true" and enabling iommu in qemu using
>> the following configuration in the guest domain XML:
>> <iommu model='intel'>
>>     <driver intremap='on'/>
>> </iommu>
>>                With iommu enabled ovs-vswitchd still crashes when guest VM restarts
>> the network service.
>>
>>                Is this a known problem? Anyone else seen a crash like this?  How can
>> we protect the ovs-vswitchd from crashing when a guest VM restarts the
>> network application or service?
>>
>> Thanks
>> Alex
>> ------------------------------------------------------------------------
>>
>> Log:
>> Oct 7 19:54:16 Branch81-Bravo kernel: [2245909.596635] pmd16[25721]:
>> segfault at 7f4d1d733000 ip 00007f4d2ae5d066 sp 00007f4d1ce65618 error 4 in
>> librte_vhost.so.4[7f4d2ae52000+1a000]
>> Oct 7 19:54:19 Branch81-Bravo systemd[1]: ovs-vswitchd.service: main process
>> exited, code=killed, status=11/SEGV
>>
>> Environment:
>> CentOs 7.6.1810
>> openvswitch-2.11.1-1.el7.centos.x86_64
>> openvswitch-kmod-2.11.1-1.el7.centos.x86_64
>> dpdk-18.11-2.el7.centos.x86_64
>> 3.10.0-1062.4.1.el7.x86_64
>> qemu-kvm-ev-2.12.0-18.el7.centos_6.1.1
>>
>> Core dump trace:
>> (gdb) bt
>> #-1 0x00007ffff205602e in rte_memcpy_generic (dst=<optimized out>,
>> src=0x7fffcef3607c, n=<optimized out>)
>> at /usr/src/debug/dpdk-18.11/x86_64-native-linuxapp-
>> gcc/include/rte_memcpy.h:793
>> Backtrace stopped: Cannot access memory at address 0x7ffff20558f0
>>
>> (gdb) list *0x00007ffff205602e
>> 0x7ffff205602e is in rte_memcpy_generic (/usr/src/debug/dpdk-18.11/x86_64-
>> native-linuxapp-gcc/include/rte_memcpy.h:793).
>> 788 }
>> 789
>> 790 /**
>> 791 * For copy with unaligned load
>> 792 */
>> 793 MOVEUNALIGNED_LEFT47(dst, src, n, srcofs);
>> 794
>> 795 /**
>> 796 * Copy whatever left
>> 797 */
>>
>> (gdb) list *0x00007ffff205c192
>> 0x7ffff205c192 is in rte_vhost_dequeue_burst (/usr/src/debug/dpdk-
>> 18.11/lib/librte_vhost/virtio_net.c:1192).
>> 1187 * In zero copy mode, one mbuf can only reference data
>> 1188 * for one or partial of one desc buff.
>> 1189 */
>> 1190 mbuf_avail = cpy_len;
>> 1191 } else {
>> 1192 if (likely(cpy_len > MAX_BATCH_LEN ||
>> 1193 vq->batch_copy_nb_elems >= vq->size ||
>> 1194 (hdr && cur == m))) {
>> 1195 rte_memcpy(rte_pktmbuf_mtod_offset(cur, void *,
>> 1196 mbuf_offset),
>> (gdb)
>>
>> _______________________________________________
>> dev mailing list
>> dev@openvswitch.org
>> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
> 

Re: [dpdk-dev] [ovs-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service

[Alex Yeh (ayeh)]

Hi Kevin, Stokes,
	Thanks for the suggestion.
	We have upgrade to OVS 2.11.4 and DPDK 18.11.10. The OVS still crashes with the same segfault error when application within the guest VM retarts. Any suggestion on how to proceed?

Thanks
Alex

[root@nfvis ~]# ovs-vswitchd --version
ovs-vswitchd (Open vSwitch) 2.11.4
DPDK 18.11.10

-----Original Message-----
From: Kevin Traynor <ktraynor@redhat.com> 
Sent: Thursday, November 19, 2020 4:09 AM
To: Stokes, Ian <ian.stokes@intel.com>; Alex Yeh (ayeh) <ayeh@cisco.com>; dev@dpdk.org
Cc: Yegappan Lakshmanan (yega) <yega@cisco.com>
Subject: Re: [dpdk-dev] [ovs-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service

On 19/11/2020 11:21, Stokes, Ian wrote:
>> Hi,
>>                We are seeing a ovs-vswitchd service crash with 
>> segfault in the librte_vhost library when a DPDK application within a guest VM is stopped.
>>
>>                We are using OVS 2.11.1 on CentOS 7.6 (3.10.0-1062 
>> Linux kernel) with DPDK 18.11.2.
> 
> Hi,
> 
> Is there a reason you are using OVS 2.11.1 and DPDK 18.11.2?  These are quite old.
> 
> As a first step I would recommend using the latest of these branches that have been validated with by the OVS community.
> 
> As of now this would be OVS 2.11.4 and DPDK 18.11.9 to check if the issue is still present there my suspicion is that this could be an issue resolved in the DPDK library since 18.11.2.
> 

+1, there's 58 commits in the vhost library on 18.11 branch since
18.11.2, so it might be already fixed. 18.11.10 is the latest release, while below is in from 18.11.7.

$ git log --oneline v18.11.2..HEAD . | grep crash 90b5ba739f vhost: fix crash on port deletion

If you are planning to continue to use 18.11 for a while, I think you will want to test the 18.11.11 Release Candidate that will be available in a few weeks. It is the last planned 18.11 release, so any issues you find *after* it is released won't be fixed.

Kevin.



> Regards
> Ian
> 
>>
>>                We are using OVS-DPDK on the host and the guest VM is 
>> running a DPDK application. With some traffic, if the application 
>> service within the VM is restarted, then OVS crashes.
>>
>>                This crash is not seen if the guest VM is restarted 
>> (instead of stopping the application within the VM).
>>
>>                The crash trackback (attached below) points to the
>> rte_memcpy_generic() function in rte_memcpy.h. It looks like the 
>> crash occurs when vhost is trying to dequeue the packets from the 
>> guest VM (as the application in the guest VM has stopped and the huge 
>> pages are returned to the guest kernel).
>>
>>                We have tried enabling iommu in ovs by setting 
>> "other_config:vhost-iommu-support=true" and enabling iommu in qemu 
>> using the following configuration in the guest domain XML:
>> <iommu model='intel'>
>>     <driver intremap='on'/>
>> </iommu>
>>                With iommu enabled ovs-vswitchd still crashes when 
>> guest VM restarts the network service.
>>
>>                Is this a known problem? Anyone else seen a crash like 
>> this?  How can we protect the ovs-vswitchd from crashing when a guest 
>> VM restarts the network application or service?
>>
>> Thanks
>> Alex
>> ---------------------------------------------------------------------
>> ---
>>
>> Log:
>> Oct 7 19:54:16 Branch81-Bravo kernel: [2245909.596635] pmd16[25721]:
>> segfault at 7f4d1d733000 ip 00007f4d2ae5d066 sp 00007f4d1ce65618 
>> error 4 in librte_vhost.so.4[7f4d2ae52000+1a000]
>> Oct 7 19:54:19 Branch81-Bravo systemd[1]: ovs-vswitchd.service: main 
>> process exited, code=killed, status=11/SEGV
>>
>> Environment:
>> CentOs 7.6.1810
>> openvswitch-2.11.1-1.el7.centos.x86_64
>> openvswitch-kmod-2.11.1-1.el7.centos.x86_64
>> dpdk-18.11-2.el7.centos.x86_64
>> 3.10.0-1062.4.1.el7.x86_64
>> qemu-kvm-ev-2.12.0-18.el7.centos_6.1.1
>>
>> Core dump trace:
>> (gdb) bt
>> #-1 0x00007ffff205602e in rte_memcpy_generic (dst=<optimized out>, 
>> src=0x7fffcef3607c, n=<optimized out>) at 
>> /usr/src/debug/dpdk-18.11/x86_64-native-linuxapp-
>> gcc/include/rte_memcpy.h:793
>> Backtrace stopped: Cannot access memory at address 0x7ffff20558f0
>>
>> (gdb) list *0x00007ffff205602e
>> 0x7ffff205602e is in rte_memcpy_generic 
>> (/usr/src/debug/dpdk-18.11/x86_64-
>> native-linuxapp-gcc/include/rte_memcpy.h:793).
>> 788 }
>> 789
>> 790 /**
>> 791 * For copy with unaligned load
>> 792 */
>> 793 MOVEUNALIGNED_LEFT47(dst, src, n, srcofs);
>> 794
>> 795 /**
>> 796 * Copy whatever left
>> 797 */
>>
>> (gdb) list *0x00007ffff205c192
>> 0x7ffff205c192 is in rte_vhost_dequeue_burst (/usr/src/debug/dpdk- 
>> 18.11/lib/librte_vhost/virtio_net.c:1192).
>> 1187 * In zero copy mode, one mbuf can only reference data
>> 1188 * for one or partial of one desc buff.
>> 1189 */
>> 1190 mbuf_avail = cpy_len;
>> 1191 } else {
>> 1192 if (likely(cpy_len > MAX_BATCH_LEN ||
>> 1193 vq->batch_copy_nb_elems >= vq->size ||
>> 1194 (hdr && cur == m))) {
>> 1195 rte_memcpy(rte_pktmbuf_mtod_offset(cur, void *,
>> 1196 mbuf_offset),
>> (gdb)
>>
>> _______________________________________________
>> dev mailing list
>> dev@openvswitch.org
>> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
> 

Re: [dpdk-dev] [ovs-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service

[Alex Yeh (ayeh)]

Hi Kevin, Stokes,
	Resending just to make sure the email is not lost.
Thanks and looking forward to your suggestion,
Alex

-----Original Message-----
From: Alex Yeh (ayeh) 
Sent: Friday, January 08, 2021 11:36 AM
To: Kevin Traynor <ktraynor@redhat.com>; Stokes, Ian <ian.stokes@intel.com>; dev@dpdk.org
Cc: Yegappan Lakshmanan (yega) <yega@cisco.com>
Subject: RE: [dpdk-dev] [ovs-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service

Hi Kevin, Stokes,
	Thanks for the suggestion.
	We have upgrade to OVS 2.11.4 and DPDK 18.11.10. The OVS still crashes with the same segfault error when application within the guest VM retarts. Any suggestion on how to proceed?

Thanks
Alex

[root@nfvis ~]# ovs-vswitchd --version
ovs-vswitchd (Open vSwitch) 2.11.4
DPDK 18.11.10

-----Original Message-----
From: Kevin Traynor <ktraynor@redhat.com>
Sent: Thursday, November 19, 2020 4:09 AM
To: Stokes, Ian <ian.stokes@intel.com>; Alex Yeh (ayeh) <ayeh@cisco.com>; dev@dpdk.org
Cc: Yegappan Lakshmanan (yega) <yega@cisco.com>
Subject: Re: [dpdk-dev] [ovs-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service

On 19/11/2020 11:21, Stokes, Ian wrote:
>> Hi,
>>                We are seeing a ovs-vswitchd service crash with 
>> segfault in the librte_vhost library when a DPDK application within a guest VM is stopped.
>>
>>                We are using OVS 2.11.1 on CentOS 7.6 (3.10.0-1062 
>> Linux kernel) with DPDK 18.11.2.
> 
> Hi,
> 
> Is there a reason you are using OVS 2.11.1 and DPDK 18.11.2?  These are quite old.
> 
> As a first step I would recommend using the latest of these branches that have been validated with by the OVS community.
> 
> As of now this would be OVS 2.11.4 and DPDK 18.11.9 to check if the issue is still present there my suspicion is that this could be an issue resolved in the DPDK library since 18.11.2.
> 

+1, there's 58 commits in the vhost library on 18.11 branch since
18.11.2, so it might be already fixed. 18.11.10 is the latest release, while below is in from 18.11.7.

$ git log --oneline v18.11.2..HEAD . | grep crash 90b5ba739f vhost: fix crash on port deletion

If you are planning to continue to use 18.11 for a while, I think you will want to test the 18.11.11 Release Candidate that will be available in a few weeks. It is the last planned 18.11 release, so any issues you find *after* it is released won't be fixed.

Kevin.



> Regards
> Ian
> 
>>
>>                We are using OVS-DPDK on the host and the guest VM is 
>> running a DPDK application. With some traffic, if the application 
>> service within the VM is restarted, then OVS crashes.
>>
>>                This crash is not seen if the guest VM is restarted 
>> (instead of stopping the application within the VM).
>>
>>                The crash trackback (attached below) points to the
>> rte_memcpy_generic() function in rte_memcpy.h. It looks like the 
>> crash occurs when vhost is trying to dequeue the packets from the 
>> guest VM (as the application in the guest VM has stopped and the huge 
>> pages are returned to the guest kernel).
>>
>>                We have tried enabling iommu in ovs by setting 
>> "other_config:vhost-iommu-support=true" and enabling iommu in qemu 
>> using the following configuration in the guest domain XML:
>> <iommu model='intel'>
>>     <driver intremap='on'/>
>> </iommu>
>>                With iommu enabled ovs-vswitchd still crashes when 
>> guest VM restarts the network service.
>>
>>                Is this a known problem? Anyone else seen a crash like 
>> this?  How can we protect the ovs-vswitchd from crashing when a guest 
>> VM restarts the network application or service?
>>
>> Thanks
>> Alex
>> ---------------------------------------------------------------------
>> ---
>>
>> Log:
>> Oct 7 19:54:16 Branch81-Bravo kernel: [2245909.596635] pmd16[25721]:
>> segfault at 7f4d1d733000 ip 00007f4d2ae5d066 sp 00007f4d1ce65618 
>> error 4 in librte_vhost.so.4[7f4d2ae52000+1a000]
>> Oct 7 19:54:19 Branch81-Bravo systemd[1]: ovs-vswitchd.service: main 
>> process exited, code=killed, status=11/SEGV
>>
>> Environment:
>> CentOs 7.6.1810
>> openvswitch-2.11.1-1.el7.centos.x86_64
>> openvswitch-kmod-2.11.1-1.el7.centos.x86_64
>> dpdk-18.11-2.el7.centos.x86_64
>> 3.10.0-1062.4.1.el7.x86_64
>> qemu-kvm-ev-2.12.0-18.el7.centos_6.1.1
>>
>> Core dump trace:
>> (gdb) bt
>> #-1 0x00007ffff205602e in rte_memcpy_generic (dst=<optimized out>, 
>> src=0x7fffcef3607c, n=<optimized out>) at
>> /usr/src/debug/dpdk-18.11/x86_64-native-linuxapp-
>> gcc/include/rte_memcpy.h:793
>> Backtrace stopped: Cannot access memory at address 0x7ffff20558f0
>>
>> (gdb) list *0x00007ffff205602e
>> 0x7ffff205602e is in rte_memcpy_generic
>> (/usr/src/debug/dpdk-18.11/x86_64-
>> native-linuxapp-gcc/include/rte_memcpy.h:793).
>> 788 }
>> 789
>> 790 /**
>> 791 * For copy with unaligned load
>> 792 */
>> 793 MOVEUNALIGNED_LEFT47(dst, src, n, srcofs);
>> 794
>> 795 /**
>> 796 * Copy whatever left
>> 797 */
>>
>> (gdb) list *0x00007ffff205c192
>> 0x7ffff205c192 is in rte_vhost_dequeue_burst (/usr/src/debug/dpdk- 
>> 18.11/lib/librte_vhost/virtio_net.c:1192).
>> 1187 * In zero copy mode, one mbuf can only reference data
>> 1188 * for one or partial of one desc buff.
>> 1189 */
>> 1190 mbuf_avail = cpy_len;
>> 1191 } else {
>> 1192 if (likely(cpy_len > MAX_BATCH_LEN ||
>> 1193 vq->batch_copy_nb_elems >= vq->size ||
>> 1194 (hdr && cur == m))) {
>> 1195 rte_memcpy(rte_pktmbuf_mtod_offset(cur, void *,
>> 1196 mbuf_offset),
>> (gdb)
>>
>> _______________________________________________
>> dev mailing list
>> dev@openvswitch.org
>> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
> 

Re: [dpdk-dev] [ovs-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service

[Kevin Traynor]

On 12/01/2021 18:20, Alex Yeh (ayeh) wrote:
> Hi Kevin, Stokes,
> 	Resending just to make sure the email is not lost.
> Thanks and looking forward to your suggestion,
> Alex
> 

+Cc vhost/virtio maintainers

Thanks for the report and checking the newer versions. I think at this
stage you should log a report in https://bugs.dpdk.org and provide steps
for the vhost/virtio maintainers so they can reproduce this issue.

> -----Original Message-----
> From: Alex Yeh (ayeh) 
> Sent: Friday, January 08, 2021 11:36 AM
> To: Kevin Traynor <ktraynor@redhat.com>; Stokes, Ian <ian.stokes@intel.com>; dev@dpdk.org
> Cc: Yegappan Lakshmanan (yega) <yega@cisco.com>
> Subject: RE: [dpdk-dev] [ovs-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service
> 
> Hi Kevin, Stokes,
> 	Thanks for the suggestion.
> 	We have upgrade to OVS 2.11.4 and DPDK 18.11.10. The OVS still crashes with the same segfault error when application within the guest VM retarts. Any suggestion on how to proceed?
> 
> Thanks
> Alex
> 
> [root@nfvis ~]# ovs-vswitchd --version
> ovs-vswitchd (Open vSwitch) 2.11.4
> DPDK 18.11.10
> 
> -----Original Message-----
> From: Kevin Traynor <ktraynor@redhat.com>
> Sent: Thursday, November 19, 2020 4:09 AM
> To: Stokes, Ian <ian.stokes@intel.com>; Alex Yeh (ayeh) <ayeh@cisco.com>; dev@dpdk.org
> Cc: Yegappan Lakshmanan (yega) <yega@cisco.com>
> Subject: Re: [dpdk-dev] [ovs-dev] ovs-vswitchd with DPDK crashed when guest VM restarts network service
> 
> On 19/11/2020 11:21, Stokes, Ian wrote:
>>> Hi,
>>>                We are seeing a ovs-vswitchd service crash with 
>>> segfault in the librte_vhost library when a DPDK application within a guest VM is stopped.
>>>
>>>                We are using OVS 2.11.1 on CentOS 7.6 (3.10.0-1062 
>>> Linux kernel) with DPDK 18.11.2.
>>
>> Hi,
>>
>> Is there a reason you are using OVS 2.11.1 and DPDK 18.11.2?  These are quite old.
>>
>> As a first step I would recommend using the latest of these branches that have been validated with by the OVS community.
>>
>> As of now this would be OVS 2.11.4 and DPDK 18.11.9 to check if the issue is still present there my suspicion is that this could be an issue resolved in the DPDK library since 18.11.2.
>>
> 
> +1, there's 58 commits in the vhost library on 18.11 branch since
> 18.11.2, so it might be already fixed. 18.11.10 is the latest release, while below is in from 18.11.7.
> 
> $ git log --oneline v18.11.2..HEAD . | grep crash 90b5ba739f vhost: fix crash on port deletion
> 
> If you are planning to continue to use 18.11 for a while, I think you will want to test the 18.11.11 Release Candidate that will be available in a few weeks. It is the last planned 18.11 release, so any issues you find *after* it is released won't be fixed.
> 
> Kevin.
> 
> 
> 
>> Regards
>> Ian
>>
>>>
>>>                We are using OVS-DPDK on the host and the guest VM is 
>>> running a DPDK application. With some traffic, if the application 
>>> service within the VM is restarted, then OVS crashes.
>>>
>>>                This crash is not seen if the guest VM is restarted 
>>> (instead of stopping the application within the VM).
>>>
>>>                The crash trackback (attached below) points to the
>>> rte_memcpy_generic() function in rte_memcpy.h. It looks like the 
>>> crash occurs when vhost is trying to dequeue the packets from the 
>>> guest VM (as the application in the guest VM has stopped and the huge 
>>> pages are returned to the guest kernel).
>>>
>>>                We have tried enabling iommu in ovs by setting 
>>> "other_config:vhost-iommu-support=true" and enabling iommu in qemu 
>>> using the following configuration in the guest domain XML:
>>> <iommu model='intel'>
>>>     <driver intremap='on'/>
>>> </iommu>
>>>                With iommu enabled ovs-vswitchd still crashes when 
>>> guest VM restarts the network service.
>>>
>>>                Is this a known problem? Anyone else seen a crash like 
>>> this?  How can we protect the ovs-vswitchd from crashing when a guest 
>>> VM restarts the network application or service?
>>>
>>> Thanks
>>> Alex
>>> ---------------------------------------------------------------------
>>> ---
>>>
>>> Log:
>>> Oct 7 19:54:16 Branch81-Bravo kernel: [2245909.596635] pmd16[25721]:
>>> segfault at 7f4d1d733000 ip 00007f4d2ae5d066 sp 00007f4d1ce65618 
>>> error 4 in librte_vhost.so.4[7f4d2ae52000+1a000]
>>> Oct 7 19:54:19 Branch81-Bravo systemd[1]: ovs-vswitchd.service: main 
>>> process exited, code=killed, status=11/SEGV
>>>
>>> Environment:
>>> CentOs 7.6.1810
>>> openvswitch-2.11.1-1.el7.centos.x86_64
>>> openvswitch-kmod-2.11.1-1.el7.centos.x86_64
>>> dpdk-18.11-2.el7.centos.x86_64
>>> 3.10.0-1062.4.1.el7.x86_64
>>> qemu-kvm-ev-2.12.0-18.el7.centos_6.1.1
>>>
>>> Core dump trace:
>>> (gdb) bt
>>> #-1 0x00007ffff205602e in rte_memcpy_generic (dst=<optimized out>, 
>>> src=0x7fffcef3607c, n=<optimized out>) at
>>> /usr/src/debug/dpdk-18.11/x86_64-native-linuxapp-
>>> gcc/include/rte_memcpy.h:793
>>> Backtrace stopped: Cannot access memory at address 0x7ffff20558f0
>>>
>>> (gdb) list *0x00007ffff205602e
>>> 0x7ffff205602e is in rte_memcpy_generic
>>> (/usr/src/debug/dpdk-18.11/x86_64-
>>> native-linuxapp-gcc/include/rte_memcpy.h:793).
>>> 788 }
>>> 789
>>> 790 /**
>>> 791 * For copy with unaligned load
>>> 792 */
>>> 793 MOVEUNALIGNED_LEFT47(dst, src, n, srcofs);
>>> 794
>>> 795 /**
>>> 796 * Copy whatever left
>>> 797 */
>>>
>>> (gdb) list *0x00007ffff205c192
>>> 0x7ffff205c192 is in rte_vhost_dequeue_burst (/usr/src/debug/dpdk- 
>>> 18.11/lib/librte_vhost/virtio_net.c:1192).
>>> 1187 * In zero copy mode, one mbuf can only reference data
>>> 1188 * for one or partial of one desc buff.
>>> 1189 */
>>> 1190 mbuf_avail = cpy_len;
>>> 1191 } else {
>>> 1192 if (likely(cpy_len > MAX_BATCH_LEN ||
>>> 1193 vq->batch_copy_nb_elems >= vq->size ||
>>> 1194 (hdr && cur == m))) {
>>> 1195 rte_memcpy(rte_pktmbuf_mtod_offset(cur, void *,
>>> 1196 mbuf_offset),
>>> (gdb)
>>>
>>> _______________________________________________
>>> dev mailing list
>>> dev@openvswitch.org
>>> https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>>
>