linux-kernel - Re: [PATCH] ARM: cpuidle: Support asymmetric idle definition

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <ce4ee3e6-baeb-5f99-a8d1-f2f3855fde35@linaro.org>
Date:   Mon, 22 May 2017 14:41:45 +0200
From:   Daniel Lezcano <daniel.lezcano@...aro.org>
To:     Sudeep Holla <sudeep.holla@....com>
Cc:     rjw@...ysocki.net, lorenzo.pieralisi@....com, leo.yan@...aro.org,
        "open list:CPUIDLE DRIVERS" <linux-pm@...r.kernel.org>,
        open list <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] ARM: cpuidle: Support asymmetric idle definition

On 22/05/2017 12:32, Sudeep Holla wrote:
> 
> 
> On 22/05/17 11:20, Daniel Lezcano wrote:
>>
>> Hi Sudeep,
>>
>>
>> On 22/05/2017 12:11, Sudeep Holla wrote:
>>>
>>>
>>> On 19/05/17 17:45, Daniel Lezcano wrote:
>>>> Some hardware have clusters with different idle states. The current code does
>>>> not support this and fails as it expects all the idle states to be identical.
>>>>
>>>> Because of this, the Mediatek mtk8173 had to create the same idle state for a
>>>> big.Little system and now the Hisilicon 960 is facing the same situation.
>>>>
>>>
>>> While I agree the we don't support them today, it's better to benchmark
>>> and record/compare the gain we get with the support for cluster based
>>> idle states.
>>
>> Sorry, I don't get what you are talking about. What do you want to
>> benchmark ? Cluster idling ?
>>
> 
> OK, I was not so clear. I had a brief chat with Lorenzo, we have few
> reason to have this support:
> 1. Different number of states between clusters
> 2. Different latencies(this is the one I was referring above, generally
>    we keep worst case timings here and wanted to see if any platform
>    measured improvements with different latencies in the idle states)

I don't see the point. Are you putting into question the big little design?

[ ... ]

>>>> +	for_each_possible_cpu(cpu) {
>>>> +
>>>> +		if (drv && cpumask_test_cpu(cpu, drv->cpumask))
>>>> +			continue;
>>>> +
>>>> +		ret = -ENOMEM;
>>>> +
>>>> +		drv = kmemdup(&arm_idle_driver, sizeof(*drv), GFP_KERNEL);
>>>> +		if (!drv)
>>>> +			goto out_fail;
>>>> +
>>>> +		drv->cpumask = &cpu_topology[cpu].core_sibling;
>>>> +
>>>
>>> This is not always true and not architecturally guaranteed. So instead
>>> of introducing this broken dependency, better to extract information
>>> from the device tree.
>>
>> Can you give an example of a broken dependency ?
>>
>> The cpu topology information is extracted from the device tree. So
>> if the topology is broken, the DT is broken also. Otherwise, the
>> topology code must fix the broken dependency from the DT.
>>
> 
> No, I meant there's no guarantee that all designs must follow this rule.
> I don't mean CPU topology code or binding is broken. What I meant is
> linking CPU topology to CPU power domains is wrong. We should make use
> of DT you infer this information as it's already there. Topology bindings
> makes no reference to power and hence you simply can't infer that
> information from it.

Ok, I will have a look how power domains can fit in this.

However I'm curious to know a platform with a cluster idle state
powering down only a subset of CPUs belonging to the cluster.


-- 
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog