[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <7ac2b77d-4633-bd3e-c24f-ec87d34b4516@gmail.com>
Date: Thu, 7 Oct 2021 00:16:11 +0300
From: Leonard Crestez <cdleonard@...il.com>
To: David Ahern <dsahern@...il.com>, Jakub Kicinski <kuba@...nel.org>,
Shuah Khan <shuah@...nel.org>, David Ahern <dsahern@...nel.org>
Cc: "David S. Miller" <davem@...emloft.net>,
Ido Schimmel <idosch@...dia.com>,
Seth David Schoen <schoen@...alty.org>,
netdev@...r.kernel.org, linux-kselftest@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 05/11] selftests: net/fcnal: kill_procs via spin instead
of sleep
On 06.10.2021 17:45, David Ahern wrote:
> On 10/6/21 5:47 AM, Leonard Crestez wrote:
>> Sleeping for one second after a kill is not necessary and adds up quite
>> quickly. Replace with a fast loop spinning until pidof returns nothing.
>>
>> Signed-off-by: Leonard Crestez <cdleonard@...il.com>
>> ---
>> tools/testing/selftests/net/fcnal-test.sh | 11 +++++++++--
>> 1 file changed, 9 insertions(+), 2 deletions(-)
>>
>> diff --git a/tools/testing/selftests/net/fcnal-test.sh b/tools/testing/selftests/net/fcnal-test.sh
>> index 0bd60cd3bc06..b7fda51deb3f 100755
>> --- a/tools/testing/selftests/net/fcnal-test.sh
>> +++ b/tools/testing/selftests/net/fcnal-test.sh
>> @@ -176,12 +176,19 @@ show_hint()
>> fi
>> }
>>
>> kill_procs()
>> {
>> - killall nettest ping ping6 >/dev/null 2>&1
>> - sleep 1
>> + local pids
>> + while true; do
>> + pids=$(pidof nettest ping ping6)
>> + if [[ -z $pids ]]; then
>> + break
>> + fi
>> + kill $pids
>> + sleep 0.01
>> + done
>> }
>>
>> do_run_cmd()
>> {
>> local cmd="$*"
>>
>
> ideally the script keeps track of processes it launches and only kills
> those. The original killall was just a stop gap until the process
> tracking was added.
That's harder to do. This is much faster and not in any way worse than
killall + sleep.
Some sort of a wrapper would have to added for each process running the
background, for each run_ping_bg.
If nettest forks by itself then $! won't work, maybe some sort of
--pid-file switch would be required?
--
Regards,
Leonard
Powered by blists - more mailing lists