[ovs-discuss] [openvswitch 2.3.2] testsuite: 717 fails on s390x

[Gerhard Stenzel]

On 10/28/2015 09:05 PM, Andy Zhou wrote:
 > On Wed, Oct 28, 2015 at 6:56 AM, Gerhard Stenzel
...
 > Thanks for pointing out this patch was missing from branch-2.3. It
 > would be nice to figure out the remaining issues (if any) before
 > cherry pick it into branch 2.3.    Thanks.
 >

When test case 717 fails, it seems the recirc_id is not as expected. So 
it is a different issue than the one fixed by 
8ae8176fd0d8ed919e3301cc961dcf02b65ff49d
When failing, the test case run takes a few seconds longer to complete.

Here is the relevant part from the log file:
...
ovs_wait
stderr:
stdout:
./ofproto-dpif.at:237: ovs-ofctl add-flows br-int flows.txt
./ofproto-dpif.at:241: ovs-appctl netdev-dummy/receive p5 
"in_port(5),eth(src=50:54:00:00:00:05,dst=50:54:00:00:01:00),eth_type(0x0800),ipv4(src=10.0.0.2,dst=10.0.0.1,proto=1,tos=0,ttl=64,frag=no)"
./ofproto-dpif.at:244: ovs-appctl upcall/disable-megaflows
./ofproto-dpif.at:247: ovs-appctl upcall/enable-megaflows
./ofproto-dpif.at:253: ovs-appctl bridge/dump-flows br1 | ofctl_strip | 
grep -- "n_packets" | grep -- "table_id" | sed -e 
's/dp_hash=0x[0-9a-f][0-9a-f]*/dp_hash=0x0/' -e 
's/output:[0-9][0-9a-f]*/output/'
--- -   2015-10-29 12:50:26.431895261 +0100
+++ 
/root/test/root/projects/ovs/tests/testsuite.dir/at-groups/717/stdout 
   2015-10-29 12:50:26.422746995 +0100
@@ -1,2 +1,2 @@
-table_id=254, n_packets=1, n_bytes=64, 
priority=20,recirc_id=0x12d,dp_hash=0x0/0xff,actions=output
+table_id=254, n_packets=1, n_bytes=64, 
priority=20,recirc_id=0x12c,dp_hash=0x0/0xff,actions=output
..