On 13 Jun 2019, at 2:37, William Tu wrote:
Hi Eelco,
#0 0x00007fbc6a78193f in raise () from /lib64/libc.so.6
#1 0x00007fbc6a76bc95 in abort () from /lib64/libc.so.6
#2 0x00000000004ed1a1 in __umem_elem_push_n
(addrs=0x7fbc40f2ec50,
n=32, umemp=0x24cc790) at lib/xdpsock.c:32
#3 umem_elem_push_n (umemp=0x24cc790, n=32,
I've found that it's due to free the afxdp twice.
The free_afxdp_buf() should be called once per dp_packet, somehow
it gets called twice.
Applying this on v11 fixes the issue
--- a/lib/dp-packet.c
+++ b/lib/dp-packet.c
@@ -145,7 +145,7 @@ dp_packet_uninit(struct dp_packet *b)
free_dpdk_buf((struct dp_packet*) b);
#endif
} else if (b->source == DPBUF_AFXDP) {
- free_afxdp_buf(b);
+ ;
}
}
}
It’s still crashing for me, even with the change. I’m using a single
nic, same port in as out:
```
@@ -122,6 +144,8 @@ dp_packet_uninit(struct dp_packet *b)
* created as a dp_packet */
free_dpdk_buf((struct dp_packet*) b);
#endif
+ } else if (b->source == DPBUF_AFXDP) {
+// free_afxdp_buf(b);
}
}
}
```
```
Core was generated by `ovs-vswitchd unix:/var/run/openvswitch/db.sock
-vconsole:emer -vsyslog:err -vfi'.
Program terminated with signal SIGABRT, Aborted.
#0 0x00007f62ef71193f in raise () from /lib64/libc.so.6
[Current thread is 1 (Thread 0x7f62bdf33700 (LWP 24737))]
Missing separate debuginfos, use: dnf debuginfo-install
elfutils-libelf-0.174-6.el8.x86_64 glibc-2.28-42.el8_0.1.x86_64
libatomic-8.2.1-3.5.el8.x86_64 libcap-ng-0.7.9-4.el8.x86_64
numactl-libs-2.0.12-2.el8.x86_64 openssl-libs-1.1.1-8.el8.x86_64
zlib-1.2.11-10.el8.x86_64
(gdb) bt
#0 0x00007f62ef71193f in raise () from /lib64/libc.so.6
#1 0x00007f62ef6fbc95 in abort () from /lib64/libc.so.6
#2 0x00000000004ed1a1 in __umem_elem_push_n (addrs=0x7f62bdf30c50,
n=32, umemp=0x2378660) at lib/xdpsock.c:32
#3 umem_elem_push_n (umemp=0x2378660, n=32,
addrs=addrs@entry=0x7f62bdf30ea0) at lib/xdpsock.c:43
#4 0x00000000009b4f41 in afxdp_complete_tx (xsk=0x2378250) at
lib/netdev-afxdp.c:736
#5 netdev_afxdp_batch_send (netdev=<optimized out>, qid=0,
batch=0x7f62a4004e80, concurrent_txq=<optimized out>) at
lib/netdev-afxdp.c:763
#6 0x0000000000908031 in netdev_send (netdev=<optimized out>,
qid=qid@entry=0, batch=batch@entry=0x7f62a4004e80,
concurrent_txq=concurrent_txq@entry=true) at lib/netdev.c:800
#7 0x00000000008d4c24 in dp_netdev_pmd_flush_output_on_port
(pmd=pmd@entry=0x7f62bdf34010, p=p@entry=0x7f62a4004e50) at
lib/dpif-netdev.c:4187
#8 0x00000000008d4f3f in dp_netdev_pmd_flush_output_packets
(pmd=pmd@entry=0x7f62bdf34010, force=force@entry=false) at
lib/dpif-netdev.c:4227
#9 0x00000000008dd2d7 in dp_netdev_pmd_flush_output_packets
(force=false, pmd=0x7f62bdf34010) at lib/dpif-netdev.c:4282
#10 dp_netdev_process_rxq_port (pmd=pmd@entry=0x7f62bdf34010,
rxq=0x237e650, port_no=1) at lib/dpif-netdev.c:4282
#11 0x00000000008dd63d in pmd_thread_main (f_=<optimized out>) at
lib/dpif-netdev.c:5449
#12 0x000000000095e94d in ovsthread_wrapper (aux_=<optimized out>) at
lib/ovs-thread.c:352
#13 0x00007f62f00312de in start_thread () from /lib64/libpthread.so.0
#14 0x00007f62ef7d6a63 in clone () from /lib64/libc.so.6
```
```
Core was generated by `ovs-vswitchd unix:/var/run/openvswitch/db.sock
-vconsole:emer -vsyslog:err -vfi'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0 netdev_afxdp_rxq_recv (rxq_=0x2bd75d0, batch=0x7f399f7fc0d0,
qfill=0x0) at lib/netdev-afxdp.c:583
583 rx->fd = xsk_socket__fd(xsk->xsk);
[Current thread is 1 (Thread 0x7f399f7fe700 (LWP 24597))]
Missing separate debuginfos, use: dnf debuginfo-install
elfutils-libelf-0.174-6.el8.x86_64 glibc-2.28-42.el8_0.1.x86_64
libatomic-8.2.1-3.5.el8.x86_64 libcap-ng-0.7.9-4.el8.x86_64
numactl-libs-2.0.12-2.el8.x86_64 openssl-libs-1.1.1-8.el8.x86_64
zlib-1.2.11-10.el8.x86_64
(gdb) bt
#0 netdev_afxdp_rxq_recv (rxq_=0x2bd75d0, batch=0x7f399f7fc0d0,
qfill=0x0) at lib/netdev-afxdp.c:583
#1 0x0000000000907f21 in netdev_rxq_recv (rx=<optimized out>,
batch=batch@entry=0x7f399f7fc0d0, qfill=<optimized out>) at
lib/netdev.c:710
#2 0x00000000008dd1c3 in dp_netdev_process_rxq_port
(pmd=pmd@entry=0x2bf80c0, rxq=0x2bd63e0, port_no=2) at
lib/dpif-netdev.c:4257
#3 0x00000000008dd63d in pmd_thread_main (f_=<optimized out>) at
lib/dpif-netdev.c:5449
#4 0x000000000095e94d in ovsthread_wrapper (aux_=<optimized out>) at
lib/ovs-thread.c:352
#5 0x00007f39d86de2de in start_thread () from /lib64/libpthread.so.0
#6 0x00007f39d7e83a63 in clone () from /lib64/libc.so.6
(gdb) p xsk->xsk
Cannot access memory at address 0x316f6ebd
(gdb) p xsk
$1 = (struct xsk_socket_info *) 0x316f6e65
```
I will work on next version
Thank you
William
<SNIP>
_______________________________________________
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
