On Wed, Jan 26, 2022 at 09:34:52PM -0800, John Fastabend wrote:
> Hangbin Liu wrote:
> > Use temp netns instead of hard code name for testing in case the netns
> > already exists.
> >
> > Remove the hard code interface index when creating the veth interfaces.
> > Because when the system loads some virtual interface modules, e.g. tunnels.
> > the ifindex of 2 will be used and the cmd will fail.
> >
> > As the netns has not created if checking environment failed. Trap the
> > clean up function after checking env.
> >
> > Fixes: 8955c1a32987 ("selftests/bpf/xdp_redirect_multi: Limit the tests in netns")
> > Signed-off-by: Hangbin Liu <liuhangbin@xxxxxxxxx>
> > ---
> > .../selftests/bpf/test_xdp_redirect_multi.sh | 60 ++++++++++---------
> > 1 file changed, 31 insertions(+), 29 deletions(-)
> >
> > diff --git a/tools/testing/selftests/bpf/test_xdp_redirect_multi.sh b/tools/testing/selftests/bpf/test_xdp_redirect_multi.sh
> > index 05f872740999..cc57cb87e65f 100755
> > --- a/tools/testing/selftests/bpf/test_xdp_redirect_multi.sh
> > +++ b/tools/testing/selftests/bpf/test_xdp_redirect_multi.sh
> > @@ -32,6 +32,11 @@ DRV_MODE="xdpgeneric xdpdrv xdpegress"
> > PASS=0
> > FAIL=0
> > LOG_DIR=$(mktemp -d)
> > +declare -a NS
> > +NS[0]="ns0-$(mktemp -u XXXXXX)"
> > +NS[1]="ns1-$(mktemp -u XXXXXX)"
> > +NS[2]="ns2-$(mktemp -u XXXXXX)"
> > +NS[3]="ns3-$(mktemp -u XXXXXX)"
> >
> > test_pass()
> > {
> > @@ -47,11 +52,9 @@ test_fail()
> >
> > clean_up()
> > {
> > - for i in $(seq $NUM); do
> > - ip link del veth$i 2> /dev/null
> > - ip netns del ns$i 2> /dev/null
> > + for i in $(seq 0 $NUM); do
> > + ip netns del ${NS[$i]} 2> /dev/null
>
> You dropped the `ip link del veth$i` why is this ok?
All the veth interfaces are created and attached in the netns. So remove
the netns should also remove related veth interfaces.
[...]
> ip -n ${NS[$i]} link add veth0 type veth peer name veth$i netns ${NS[0]}
[...]
Thanks
Hangbin
