From: Matthias Brugger <matthias.bgg@gmail.com>
To: AngeloGioacchino Del Regno
<angelogioacchino.delregno@collabora.com>,
robh+dt@kernel.org
Cc: krzysztof.kozlowski+dt@linaro.org, mturquette@baylibre.com,
sboyd@kernel.org, p.zabel@pengutronix.de,
y.oudjana@protonmail.com, jason-jh.lin@mediatek.com,
ck.hu@mediatek.com, fparent@baylibre.com,
rex-bc.chen@mediatek.com, tinghan.shen@mediatek.com,
chun-jie.chen@mediatek.com, weiyi.lu@mediatek.com,
ikjn@chromium.org, miles.chen@mediatek.com,
sam.shih@mediatek.com, wenst@chromium.org,
bgolaszewski@baylibre.com, devicetree@vger.kernel.org,
linux-kernel@vger.kernel.org,
linux-arm-kernel@lists.infradead.org,
linux-mediatek@lists.infradead.org, linux-clk@vger.kernel.org,
konrad.dybcio@somainline.org, marijn.suijten@somainline.org,
martin.botka@somainline.org,
~postmarketos/upstreaming@lists.sr.ht,
phone-devel@vger.kernel.org, paul.bouchara@somainline.org,
kernel@collabora.com
Subject: Re: [PATCH 5/5] clk: mediatek: Add MediaTek Helio X10 MT6795 clock drivers
Date: Tue, 17 May 2022 10:48:04 +0200 [thread overview]
Message-ID: <739b3ccc-e8ac-2748-9d40-0666a4038876@gmail.com> (raw)
In-Reply-To: <b65546e1-2d3a-d525-b664-1730dc06994f@collabora.com>
Hi Angelo,
On 17/05/2022 10:07, AngeloGioacchino Del Regno wrote:
> Il 16/05/22 13:30, Matthias Brugger ha scritto:
>>
>>
>> On 13/05/2022 18:50, AngeloGioacchino Del Regno wrote:
>>> Add the clock drivers for the entire clock tree of MediaTek Helio X10
>>> MT6795, including system clocks (apmixedsys, infracfg, pericfg, topckgen)
>>> and multimedia clocks (mmsys, mfg, vdecsys, vencsys).
>>>
>>> Signed-off-by: AngeloGioacchino Del Regno
>>> <angelogioacchino.delregno@collabora.com>
>>
>> Thanks a lot for taking care of this!
>> I just wonder if we couldn't build most of the clock drivers as modules like
>> done for the mt6779. It would help us to keep the kernel image smaller.
>>
>
> Hello Matthias!
>
> You're welcome!
> ...but I simply couldn't stand at seeing partially working (..or actually, not
> really working) SoCs upstream. If something is upstream, it must work, or it
> shouldn't be here for real :-)
>
> Regarding your question about the clock drivers as module... I believe we can,
> but that'd be only for {vdec,venc}sys and *maybe* MFG (gpu clocks): I don't know
> if it'd be worth to do, as these are about... 8 clocks out of... I haven't counted
> them, but more than 250, I think?
>
> It *should* be straightforward though, just about giving them a tristate in Kconfig
> instead of a bool, but that would still be limited to just those three...
>
I think you could guide yourself by looking at
f09b9460a5e4 ("clk: mediatek: support COMMON_CLK_MT6779 module build")
I think having clocks as modules is a criteria to be part of Android's Generic
Kernel Image. Not that we target this here, just for your information (in case
you didn't know).
> The reason for me excluding clk-mt6795-mm from this choice is that - at least for
> me - my development platform is a commercial smartphone, where the only thing that
> "saves you" is having some display output... I mean - I *do* have a UART port, but
> that's only because I've been able to solder thin wires on 0.2mm pads... you surely
> agree on the fact that this isn't a common practice, even across developers.
>
Yes I was wondering if you got some development smartphone or you did the
soldering. I have some solder knowledge but I think not enough to solder
something like this :D
> Besides, if you think that clk-mt6795-mm should indeed be a module by default,
> well, that.. is.. possible - I don't see why it shouldn't be... obviously keeping
> in mind that this will largely slow down the boot process, which isn't a big issue.
>
Well I don't want to bother you with minor details, most important thing is that
we get support upstream... but ;) you can always mark the modules as compiled
in, even if it's tristate, which in your case you want to do.
> In any case, it is *not* possible to compile as module *any* of the clock drivers
> that I have included in the CONFIG_COMMON_CLK_MT6795 (apmixed, infra, peri, topck)
> as.. you know.. these are "a bit critical" on older platforms :-)
>
>
> How would you proceed?
>
I'd make all the drivers that theoretically can be build as modules tristate,
you can then define for your development environment to make them build-in.
Sounds good?
Regards,
Matthias
next prev parent reply other threads:[~2022-05-17 8:48 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-13 16:50 [PATCH 0/5] MediaTek Helio X10 MT6795 - Clock drivers AngeloGioacchino Del Regno
2022-05-13 16:50 ` [PATCH 1/5] dt-bindings: mediatek: Document MT6795 system controllers bindings AngeloGioacchino Del Regno
2022-05-16 16:03 ` Rob Herring
2022-05-13 16:50 ` [PATCH 2/5] dt-bindings: clock: Add MediaTek Helio X10 MT6795 clock bindings AngeloGioacchino Del Regno
2022-05-13 16:50 ` [PATCH 3/5] dt-bindings: reset: Add bindings for MT6795 Helio X10 reset controllers AngeloGioacchino Del Regno
2022-05-13 16:50 ` [PATCH 4/5] dt-bindings: arm: mediatek: Add clock driver bindings for MT6795 AngeloGioacchino Del Regno
2022-05-13 20:48 ` Rob Herring
2022-05-16 17:28 ` Rob Herring
2022-05-17 7:48 ` AngeloGioacchino Del Regno
2022-05-13 16:50 ` [PATCH 5/5] clk: mediatek: Add MediaTek Helio X10 MT6795 clock drivers AngeloGioacchino Del Regno
2022-05-13 22:00 ` kernel test robot
2022-05-13 23:01 ` kernel test robot
2022-05-16 11:30 ` Matthias Brugger
2022-05-17 8:07 ` AngeloGioacchino Del Regno
2022-05-17 8:48 ` Matthias Brugger [this message]
2022-05-17 8:59 ` AngeloGioacchino Del Regno
2022-05-13 16:54 ` [PATCH 0/5] MediaTek Helio X10 MT6795 - Clock drivers AngeloGioacchino Del Regno
2022-05-18 0:47 ` Rob Herring
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=739b3ccc-e8ac-2748-9d40-0666a4038876@gmail.com \
--to=matthias.bgg@gmail.com \
--cc=angelogioacchino.delregno@collabora.com \
--cc=bgolaszewski@baylibre.com \
--cc=chun-jie.chen@mediatek.com \
--cc=ck.hu@mediatek.com \
--cc=devicetree@vger.kernel.org \
--cc=fparent@baylibre.com \
--cc=ikjn@chromium.org \
--cc=jason-jh.lin@mediatek.com \
--cc=kernel@collabora.com \
--cc=konrad.dybcio@somainline.org \
--cc=krzysztof.kozlowski+dt@linaro.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-clk@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mediatek@lists.infradead.org \
--cc=marijn.suijten@somainline.org \
--cc=martin.botka@somainline.org \
--cc=miles.chen@mediatek.com \
--cc=mturquette@baylibre.com \
--cc=p.zabel@pengutronix.de \
--cc=paul.bouchara@somainline.org \
--cc=phone-devel@vger.kernel.org \
--cc=rex-bc.chen@mediatek.com \
--cc=robh+dt@kernel.org \
--cc=sam.shih@mediatek.com \
--cc=sboyd@kernel.org \
--cc=tinghan.shen@mediatek.com \
--cc=weiyi.lu@mediatek.com \
--cc=wenst@chromium.org \
--cc=y.oudjana@protonmail.com \
--cc=~postmarketos/upstreaming@lists.sr.ht \
/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;
as well as URLs for NNTP newsgroup(s).