Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH 11/11] selftests: net/fcnal: Reduce client timeout
To:     Leonard Crestez <cdleonard@gmail.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Shuah Khan <shuah@kernel.org>, David Ahern <dsahern@kernel.org>
Cc:     "David S. Miller" <davem@davemloft.net>,
        Ido Schimmel <idosch@nvidia.com>,
        Seth David Schoen <schoen@loyalty.org>,
        netdev@vger.kernel.org, linux-kselftest@vger.kernel.org,
        linux-kernel@vger.kernel.org
References: <cover.1633520807.git.cdleonard@gmail.com>
 <516043441bd13bc1e6ba7f507a04362e04c06da5.1633520807.git.cdleonard@gmail.com>
 <3ed2262e-fce2-c587-5112-e4583cd042ed@gmail.com>
 <c48ea9e2-acdc-eb11-a4b0-35474003fcf3@gmail.com>
From:   David Ahern <dsahern@gmail.com>
Message-ID: <65ae97e3-73c1-3221-96fe-6096a8aacfa1@gmail.com>
Date:   Wed, 6 Oct 2021 19:17:47 -0600
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:78.0)
 Gecko/20100101 Thunderbird/78.14.0
MIME-Version: 1.0
In-Reply-To: <c48ea9e2-acdc-eb11-a4b0-35474003fcf3@gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 10/6/21 3:26 PM, Leonard Crestez wrote:
> 
> 
> On 06.10.2021 18:01, David Ahern wrote:
>> On 10/6/21 5:47 AM, Leonard Crestez wrote:
>>> Reduce default client timeout from 5 seconds to 500 miliseconds.
>>> Can be overridden from environment by exporting NETTEST_CLIENT_TIMEOUT=5
>>>
>>> Some tests need ICMP timeouts so pass an explicit -t5 for those.
>>>
>>> Signed-off-by: Leonard Crestez <cdleonard@gmail.com>
>>> ---
>>>   tools/testing/selftests/net/fcnal-test.sh | 17 +++++++++++------
>>>   1 file changed, 11 insertions(+), 6 deletions(-)
>>>
>>
>> The problem with blindly reducing the timeouts is running the script on
>> a loaded server. Some tests are expected to timeout while for tests a
>> timeout is a failure.
> 
> Keeping the default value "5" would be fine as long as it is possible to
> override externally and get fast results on a mostly-idle machine.

5 is the default for nettest.c; the test script passes in -t1 for all tests.

> 
> Placing a default value in the environment which is overriden by certain
> tests achieves that.
> 
> In theory it would also be possible for fcnal-test.sh to parse as
> "--timeout" option and pass it into every single test but that solution
> would cause much more code churn.
> 
> Having default values in environment variables that can still be
> overridden by command-line arguments is a common pattern in many tools.
> It also avoids having to pass-through every flag through every
> intermediate wrapper.

I do not agree with env variables here.