Re: [dpdk-dev] [PATCH] app/test: enhance sanity script

["Burakov, Anatoly" <anatoly.burakov@intel.com>]

On 03-May-18 11:03 AM, Parthasarathy, JananeeX M wrote:
> Hi Anatoly,
> 
> As per the comment " Since file prefixes are not supported, tests cannot be run in parallel"
> Does this means we need not run parallel test group test cases for FreeBSD. So the total test count for FreeBSD will be based on Non-Parallel test groups only.

No, the tests in the "parallel group" can be run in parallel, but on 
FreeBSD you cannot run them *in parallel* - that doesn't mean you don't 
have to run them :) Just ensure that the thread pool size is set to 1 on 
FreeBSD.

> 
> Please confirm / share your views on the same.
> 
> Thanks
> M.P.Jananee
> 
> -----Original Message-----
> From: Burakov, Anatoly
> Sent: Friday, April 27, 2018 7:26 PM
> To: Pattan, Reshma <reshma.pattan@intel.com>; Mcnamara, John <john.mcnamara@intel.com>; dev@dpdk.org
> Cc: Parthasarathy, JananeeX M <jananeex.m.parthasarathy@intel.com>
> Subject: Re: [dpdk-dev] [PATCH] app/test: enhance sanity script
> 
> On 27-Apr-18 2:20 PM, Reshma Pattan wrote:
>> 1. Sanity Script is enhanced with additional test cases being added to
>> autotest_data.py 2. Fixed in autotest_test_funcs.py to handle test
>> cases which returns "Skipped" as result.
>>     The issue was skipped test cases got timed out, causing delay in
>> sanity script execution.
>> 3. Enhanced support for FreeBSD to add the respective memory limit, as
>> FreeBSD doesn't support socket-mem, file-prefix options.
>>
>> Signed-off-by: Jananee Parthasarathy
>> <jananeex.m.parthasarathy@intel.com>
>> ---
> 
> <...>
> 
>> @@ -1,5 +1,5 @@
>>    # SPDX-License-Identifier: BSD-3-Clause -# Copyright(c) 2010-2014
>> Intel Corporation
>> +# Copyright(c) 2010-2018 Intel Corporation
>>    
>>    # The main logic behind running autotests in parallel
>>    
>> @@ -201,6 +201,13 @@ def __init__(self, cmdline, target, blacklist, whitelist):
>>        # set up cmdline string
>>        def __get_cmdline(self, test):
>>            cmdline = self.cmdline
>> +        os_name = platform.system()
>> +
>> +        # append the respective memory limitation on FREEBSD
>> +        # as socket-mem, file-prefix is not supported
>> +        if os_name == "FreeBSD" :
>> +            cmdline += " -m " + str(sum(map(int, test["Memory"].split(","))))
>> +            return cmdline
> 
> A better fix would've been to completely avoid calling __get_cmdline() for FreeBSD. Since file prefixes are not supported, tests cannot be run in parallel, so there is no need to limit allocated memory in the first place, so neither -m nor --socket-mem switches need to be added either.
> And without those, there's no need to modify the test cmdline at all.

Also, regarding this - i think what i suggested earlier is slightly 
incorrect. A better fix would be to simply return cmdline without 
modifications on FreeBSD. Just because we don't modify cmdline now 
doesn't mean we wouldn't be doing so in the future, so not calling 
__get_cmdline is not the best solution.

> 
> --
> Thanks,
> Anatoly
> 


-- 
Thanks,
Anatoly