Re: [PATCH 17/24] selftests/resctrl: Create struct for input parameter

From: Reinette Chatre <reinette.chatre@intel.com>
To: "Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>
Cc: linux-kselftest@vger.kernel.org, "Shuah Khan" <shuah@kernel.org>,
	"Shaopeng Tan" <tan.shaopeng@jp.fujitsu.com>,
	"Maciej Wieczór-Retman" <maciej.wieczor-retman@intel.com>,
	"Fenghua Yu" <fenghua.yu@intel.com>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 17/24] selftests/resctrl: Create struct for input parameter
Date: Fri, 3 Nov 2023 15:50:34 -0700	[thread overview]
Message-ID: <ace11119-8de6-4e68-a966-431ccfa14b9d@intel.com> (raw)
In-Reply-To: <2e266bb-653f-2fe2-9dbc-db8388f6aff1@linux.intel.com>

Hi Ilpo,

On 11/3/2023 4:24 AM, Ilpo Järvinen wrote:
> On Thu, 2 Nov 2023, Reinette Chatre wrote:
>> On 10/24/2023 2:26 AM, Ilpo Järvinen wrote:
>>
>>> diff --git a/tools/testing/selftests/resctrl/mba_test.c b/tools/testing/selftests/resctrl/mba_test.c
>>> index d3bf4368341e..5157a3f74fee 100644
>>> --- a/tools/testing/selftests/resctrl/mba_test.c
>>> +++ b/tools/testing/selftests/resctrl/mba_test.c
>>> @@ -141,13 +141,13 @@ void mba_test_cleanup(void)
>>>  	remove(RESULT_FILE_NAME);
>>>  }
>>>  
>>> -int mba_schemata_change(int cpu_no, const char * const *benchmark_cmd)
>>> +int mba_schemata_change(const struct user_params *uparams)
>>>  {
>>>  	struct resctrl_val_param param = {
>>>  		.resctrl_val	= MBA_STR,
>>>  		.ctrlgrp	= "c1",
>>>  		.mongrp		= "m1",
>>> -		.cpu_no		= cpu_no,
>>> +		.cpu_no		= uparams->cpu,
>>>  		.filename	= RESULT_FILE_NAME,
>>>  		.bw_report	= "reads",
>>>  		.setup		= mba_setup
>>> @@ -156,7 +156,7 @@ int mba_schemata_change(int cpu_no, const char * const *benchmark_cmd)
>>>  
>>>  	remove(RESULT_FILE_NAME);
>>>  
>>> -	ret = resctrl_val(benchmark_cmd, &param);
>>> +	ret = resctrl_val(uparams->benchmark_cmd, &param);
>>>  	if (ret)
>>>  		goto out;
>>>  
>>
>> How about a new member of struct resctrl_val_param that points to 
>> uparams? That would remove cpu_no from resctrl_val_param
>> and have everything available when a test needs to run ... not copying
>> some user parameters into struct resctrl_val_param and passing
>> others as parameters.
> 
> I'm a bit allergic to adding more stuff into resctrl_val_param. It seems 
> a structure where random stuff has been thrown at just because it exists.
> In general, your point is very valid though because the members of 
> resctrl_val_param should be auditted through to see how many of them are 
> even useful after adding uparams and struct resctrl_test.
> 
> I could get rid of copying parameters from uparams to params and just 
> passing uparams instead of benchmark_cmd into resctrl_val(). Would you be 
> okay with that?

I am ok with that. I assume this implies that cpu_no will be removed from
resctrl_val_param?

> Oh, and I really should rename resctrl_val() one day to something more 
> meaningful too. :-) (but it won't be part of this series and will likely 
> be another conflicty nightmare because resctrl_val_param too needs to 
> be renamed...).

"Naming only" changes that are not part of something more substantive are not
very appealing though.

Reinette