mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Chun-Jen Tseng (曾俊仁)" <Chun-Jen.Tseng@mediatek.com>
To: "viresh.kumar@linaro.org" <viresh.kumar@linaro.org>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-mediatek@lists.infradead.org" 
	<linux-mediatek@lists.infradead.org>,
	"devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
	"linux-pm@vger.kernel.org" <linux-pm@vger.kernel.org>,
	"sumitg@nvidia.com" <sumitg@nvidia.com>,
	"sanjayc@nvidia.com" <sanjayc@nvidia.com>,
	"rafael.j.wysocki@intel.com" <rafael.j.wysocki@intel.com>,
	Project_Global_Chrome_Upstream_Group 
	<Project_Global_Chrome_Upstream_Group@mediatek.com>,
	"linux-arm-kernel@lists.infradead.org" 
	<linux-arm-kernel@lists.infradead.org>,
	"matthias.bgg@gmail.com" <matthias.bgg@gmail.com>,
	"rafael@kernel.org" <rafael@kernel.org>,
	"angelogioacchino.delregno@collabora.com" 
	<angelogioacchino.delregno@collabora.com>
Subject: Re: [PATCH] cpufreq: mediatek: change transition delay for MT8186
Date: Tue, 29 Aug 2023 06:57:01 +0000	[thread overview]
Message-ID: <86f49ce84f32941185c961da8a5c671e7aed46b1.camel@mediatek.com> (raw)
In-Reply-To: <20230828063904.r7huxclehlblkkjx@vireshk-i7>

On Mon, 2023-08-28 at 12:09 +0530, Viresh Kumar wrote:
>  	 
> External email : Please do not click links or open attachments until
> you have verified the sender or the content.
>  Hi Mark,
> 
> I am not entirely clear by few things in the commit log.
> 
> On 18-08-23, 10:06, Mark Tseng wrote:
> > For MT8186, it has policy0 and policy6 by different governor
> thread,so
> > it may be call cpufreq->set_target_index() by different core.
> 
> Why does this matter ?

For MT8186 set freq must Big CPU -> Little CPU -> CCI like this order
and it 
takes 10 ms.

But in cpufreq & devfreq flow , when Big CPU or Little CPU change freq
, it will call
CCI setting by different policy. It will become Big CPU -> CCI setting
or Little CPU ->
CCI setting. Howevery, It will cause CCI setting to wrong value when
per 1ms call governor
and change freq.

> > In general
> > case, it must check BCPU, LCPU and CCI together then take about
> 10ms.
> 
> BCPU is Big CPU ? LCPU is Little CPU ?

Yes, it is.

> So are you saying that changing the frequency takes roughly 10 ms for
> MT8186 ?
> 
> > Atfer 44295af5019f this patch, it may call cpufreq_out_of_sync() by
> > cpufreq_verify_current_freq() because current frequency is bigger
> > than clk_get_rate() over 1 Mhz. By the same time, it may call
> 
> s/ouver/over/
> s/1Mh/1 MHz/
> 
> 
> > cpufreq->set_target_index() again.
> 
> Where was it called for the first time ?

this patch (44295af5019f) , fixed cpufreq_out_of_sync() condition from
1000 Mhz to 1 Mhz.
So, when cpufreq_verify_current_freq() call clk_get_rate() over 1 Mhz,
it must to do freq sync
by cpufreq_out_of_sync(). In MT8186, it offen over 1 Mhz when call
clk_get_rate(), so I modify
transition delay from 1 ms to 10 ms for make sure freq setting done. 

> > So, the CCI freq may be too lower for
> > BCPU cause BCPU kernel panic.
> 
> I am not sure how a low frequency causes kernel panic here.

In MT8186, if CCI freq is lower than Bit CPU freq 70%, it will causes
kernel panic
on Big CPU.

> > So, it should change the default transition delay 1ms to 10ms. It
> can
> > promise the next freq setting then governor trigger new freq
> change.
> 
> There are few typos as well here, please fix them.
> 
> > Fixes: 44295af5019f ("cpufreq: use correct unit when verify cur
> freq")
> 
> I think you should drop this. The issue at hand may be visible now
> after 44295af5019f is applied, but it certainly didn't cause it.
> 
> -- 
> viresh
> 

  reply	other threads:[~2023-08-29  6:58 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-18  2:06 Mark Tseng
2023-08-28  6:39 ` Viresh Kumar
2023-08-29  6:57   ` Chun-Jen Tseng (曾俊仁) [this message]
2023-08-29  7:10     ` Viresh Kumar
2023-08-29  8:25       ` Chun-Jen Tseng (曾俊仁)
2023-08-29  8:31         ` Viresh Kumar
2023-08-29  8:35           ` Viresh Kumar

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=86f49ce84f32941185c961da8a5c671e7aed46b1.camel@mediatek.com \
    --to=chun-jen.tseng@mediatek.com \
    --cc=Project_Global_Chrome_Upstream_Group@mediatek.com \
    --cc=angelogioacchino.delregno@collabora.com \
    --cc=devicetree@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mediatek@lists.infradead.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=matthias.bgg@gmail.com \
    --cc=rafael.j.wysocki@intel.com \
    --cc=rafael@kernel.org \
    --cc=sanjayc@nvidia.com \
    --cc=sumitg@nvidia.com \
    --cc=viresh.kumar@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®