netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Kui-Feng Lee <sinquersw@gmail.com>
To: Paolo Abeni <pabeni@redhat.com>,
	kuifeng@meta.com, dsahern@kernel.org, davem@davemloft.net,
	edumazet@google.com, kuba@kernel.org, netdev@vger.kernel.org,
	martin.lau@linux.dev, kernel-team@meta.com, yhs@meta.com
Cc: thinker.li@gmail.com
Subject: Re: [PATCH net-next v4 2/2] selftests: fib_tests: Add a test case for IPv6 garbage collection
Date: Tue, 1 Aug 2023 15:39:02 -0700	[thread overview]
Message-ID: <bbc002f6-2f76-aa85-edfb-a1764d28dc11@gmail.com> (raw)
In-Reply-To: <bcb2066d-465a-6d2a-8469-7d60adc3f276@gmail.com>



On 7/31/23 11:14, Kui-Feng Lee wrote:
> Sorry for replying late! I am just back from a vocation.
> 
> On 7/25/23 03:27, Paolo Abeni wrote:
>> On Fri, 2023-07-21 at 17:38 -0700, kuifeng@meta.com wrote:
>>> From: Kui-Feng Lee <kuifeng@meta.com>
>>>
>>> Add 10 IPv6 routes with expiration time.  Wait for a few seconds
>>> to make sure they are removed correctly.
>>>
>>> Signed-off-by: Kui-Feng Lee <kuifeng@meta.com>
>>> ---
>>>   tools/testing/selftests/net/fib_tests.sh | 90 +++++++++++++++++++++++-
>>>   1 file changed, 87 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/tools/testing/selftests/net/fib_tests.sh 
>>> b/tools/testing/selftests/net/fib_tests.sh
>>> index 35d89dfa6f11..4c92fb3c3844 100755
>>> --- a/tools/testing/selftests/net/fib_tests.sh
>>> +++ b/tools/testing/selftests/net/fib_tests.sh
>>> @@ -9,13 +9,16 @@ ret=0
>>>   ksft_skip=4
>>>   # all tests in this script. Can be overridden with -t option
>>> -TESTS="unregister down carrier nexthop suppress ipv6_notify 
>>> ipv4_notify ipv6_rt ipv4_rt ipv6_addr_metric ipv4_addr_metric 
>>> ipv6_route_metrics ipv4_route_metrics ipv4_route_v6_gw rp_filter 
>>> ipv4_del_addr ipv4_mangle ipv6_mangle ipv4_bcast_neigh"
>>> +TESTS="unregister down carrier nexthop suppress ipv6_notify 
>>> ipv4_notify \
>>> +       ipv6_rt ipv4_rt ipv6_addr_metric ipv4_addr_metric 
>>> ipv6_route_metrics \
>>> +       ipv4_route_metrics ipv4_route_v6_gw rp_filter ipv4_del_addr \
>>> +       ipv4_mangle ipv6_mangle ipv4_bcast_neigh fib6_gc_test"
>>>   VERBOSE=0
>>>   PAUSE_ON_FAIL=no
>>>   PAUSE=no
>>> -IP="ip -netns ns1"
>>> -NS_EXEC="ip netns exec ns1"
>>> +IP="$(which ip) -netns ns1"
>>> +NS_EXEC="$(which ip) netns exec ns1"
>>>   which ping6 > /dev/null 2>&1 && ping6=$(which ping6) || 
>>> ping6=$(which ping)
>>> @@ -747,6 +750,86 @@ fib_notify_test()
>>>       cleanup &> /dev/null
>>>   }
>>> +fib6_gc_test()
>>> +{
>>> +    setup
>>> +
>>> +    echo
>>> +    echo "Fib6 garbage collection test"
>>> +    set -e
>>> +
>>> +    # Check expiration of routes every 3 seconds (GC)
>>> +    $NS_EXEC sysctl -wq net.ipv6.route.gc_interval=300
>>> +
>>> +    $IP link add dummy_10 type dummy
>>> +    $IP link set dev dummy_10 up
>>> +    $IP -6 address add 2001:10::1/64 dev dummy_10
>>> +
>>> +    $NS_EXEC sysctl -wq net.ipv6.route.flush=1
>>> +
>>> +    # Temporary routes
>>> +    for i in $(seq 1 1000); do
>>> +        # Expire route after 4 seconds
>>> +        $IP -6 route add 2001:20::$i \
>>> +        via 2001:10::2 dev dummy_10 expires 4
>>> +    done
>>> +    N_EXP=$($IP -6 route list |grep expires|wc -l)
>>> +    if [ $N_EXP -ne 1000 ]; then
>>> +        echo "FAIL: expected 1000 routes with expires, got $N_EXP"
>>> +        ret=1
>>> +    else
>>> +        sleep 5
>>> +        REALTM_P=$($NS_EXEC strace -T sysctl \
>>> +               -wq net.ipv6.route.flush=1 2>&1 | \
>>> +               awk -- '/write\(.*"1\\n", 2\)/ { gsub("(.*<|>.*)", 
>>> ""); print $0;}')
>>
>> I guess the above works somehow ?!?
>>
>> But I think something alike:
>>
>>     # just after printing the banner
>>     TIME=$(which time)
>>     if [ -z "$TIME" ]; then
>>         echo "command 'time' is missing, skipping test"
>>         return
>>     fi
>>     # ...
>>
>>         # replacing the strace command
>>         REALTM_P=$(time -f %e $NS_EXEC sysctl \
>>                -wq net.ipv6.route.flush=1 2>&1)
>>
>> would be better.
>>
>> In any case you should check explicitly for the additionally needed
>> command ('strace' in your code, 'time' here).
>>
>> And you could include the expected output in the commit message (just a
>> line, right?)
>>
>> Cheers
>>
>> Paolo
> 
> 
> The availability of 'time' is much higher than 'strace', however
> 'time -f %e' measures runtime in mini-seconds. That means I have to
> add more (x100 perhaps) temporary routes to make it reliably visible to
> 'time -f %e'. I will try it and check the existence of the command.

It is really slow adding 100000 routes. I turned to 'strace' and
detecting its existence.

> 
>>
>>
>>> +        N_EXP_s5=$($IP -6 route list |grep expires|wc -l)
>>> +
>>> +        if [ $N_EXP_s5 -ne 0 ]; then
>>> +        echo "FAIL: expected 0 routes with expires, got $N_EXP_s5"
>>> +        ret=1
>>> +        else
>>> +        ret=0
>>> +        fi
>>> +    fi
>>> +
>>> +    # Permanent routes
>>> +    for i in $(seq 1 5000); do
>>> +        $IP -6 route add 2001:30::$i \
>>> +        via 2001:10::2 dev dummy_10
>>> +    done
>>> +    # Temporary routes
>>> +    for i in $(seq 1 1000); do
>>> +        # Expire route after 4 seconds
>>> +        $IP -6 route add 2001:20::$i \
>>> +        via 2001:10::2 dev dummy_10 expires 4
>>> +    done
>>> +    N_EXP=$($IP -6 route list |grep expires|wc -l)
>>> +    if [ $N_EXP -ne 1000 ]; then
>>> +        echo
>>> +        "FAIL: expected 1000 routes with expires, got $N_EXP (5000 
>>> permanent routes)"
>>> +        ret=1
>>> +    else
>>> +        sleep 5
>>> +        REALTM_T=$($NS_EXEC strace -T sysctl \
>>> +               -wq net.ipv6.route.flush=1 2>&1 | \
>>> +               awk -- '/write\(.*"1\\n", 2\)/ { gsub("(.*<|>.*)", 
>>> ""); print $0;}')
>>> +        N_EXP_s5=$($IP -6 route list |grep expires|wc -l)
>>> +
>>> +        if [ $N_EXP_s5 -ne 0 ]; then
>>> +        echo "FAIL: expected 0 routes with expires, got $N_EXP_s5 
>>> (5000 permanent routes)"
>>> +        ret=1
>>> +        else
>>> +        ret=0
>>> +        fi
>>> +    fi
>>> +
>>> +    set +e
>>> +
>>> +    log_test $ret 0 "ipv6 route garbage collection (${REALTM_P}s, 
>>> ${REALTM_T}s)"
>>> +
>>> +    cleanup &> /dev/null
>>> +}
>>> +
>>>   fib_suppress_test()
>>>   {
>>>       echo
>>> @@ -2217,6 +2300,7 @@ do
>>>       ipv4_mangle)            ipv4_mangle_test;;
>>>       ipv6_mangle)            ipv6_mangle_test;;
>>>       ipv4_bcast_neigh)        ipv4_bcast_neigh_test;;
>>> +    fib6_gc_test|ipv6_gc)        fib6_gc_test;;
>>>       help) echo "Test names: $TESTS"; exit 0;;
>>>       esac
>>
>>

      reply	other threads:[~2023-08-01 22:39 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-22  0:38 [PATCH net-next v4 0/2] Remove expired routes with a separated list of routes kuifeng
2023-07-22  0:38 ` [PATCH net-next v4 1/2] net/ipv6: " kuifeng
2023-07-22  0:38 ` [PATCH net-next v4 2/2] selftests: fib_tests: Add a test case for IPv6 garbage collection kuifeng
2023-07-25 10:27   ` Paolo Abeni
2023-07-31 18:14     ` Kui-Feng Lee
2023-08-01 22:39       ` Kui-Feng Lee [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=bbc002f6-2f76-aa85-edfb-a1764d28dc11@gmail.com \
    --to=sinquersw@gmail.com \
    --cc=davem@davemloft.net \
    --cc=dsahern@kernel.org \
    --cc=edumazet@google.com \
    --cc=kernel-team@meta.com \
    --cc=kuba@kernel.org \
    --cc=kuifeng@meta.com \
    --cc=martin.lau@linux.dev \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=thinker.li@gmail.com \
    --cc=yhs@meta.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).