linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Thierry Reding <thierry.reding@gmail.com>
To: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Cc: Jon Hunter <jonathanh@nvidia.com>,
	Dmitry Osipenko <digetx@gmail.com>,
	linux-tegra@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH 03/10] memory: tegra: Push suspend/resume into SoC drivers
Date: Mon, 26 Apr 2021 14:24:46 +0200	[thread overview]
Message-ID: <YIaxDk39hjMJ6nxA@orome.fritz.box> (raw)
In-Reply-To: <d1626fd2-90e3-7a26-51b8-497a344aee6f@canonical.com>


[-- Attachment #1.1: Type: text/plain, Size: 6006 bytes --]

On Mon, Apr 26, 2021 at 10:47:11AM +0200, Krzysztof Kozlowski wrote:
> On 20/04/2021 18:52, Thierry Reding wrote:
> > From: Thierry Reding <treding@nvidia.com>
> > 
> > Continuing the scheme of unification, push suspend/resume callbacks into
> > per-SoC driver so that they can be properly parameterized.
> > 
> > While at it, also move the ->init() callback into the new tegra_mc_ops
> > structure to keep things clean.
> 
> Please split this part. This is just moving pointer from one structure
> to another, quite small change.
> 
> The rest of the patchset is quite different - you now call
> tegra_mc_suspend() from a per-SoC driver and move the code around.

Okay, I think I can split that off into a separate patch.

> > Signed-off-by: Thierry Reding <treding@nvidia.com>
> > ---
> >  drivers/memory/tegra/mc.c       | 24 ++++++++---------------
> >  drivers/memory/tegra/tegra186.c | 27 ++++++++++++++++++++++----
> >  drivers/memory/tegra/tegra20.c  | 34 ++++++++++++++++++++++++++++++++-
> >  include/soc/tegra/mc.h          |  9 +++++++--
> >  4 files changed, 71 insertions(+), 23 deletions(-)
> > 
> > diff --git a/drivers/memory/tegra/mc.c b/drivers/memory/tegra/mc.c
> > index b7e104bf6614..2b21131d779c 100644
> > --- a/drivers/memory/tegra/mc.c
> > +++ b/drivers/memory/tegra/mc.c
> > @@ -829,8 +829,8 @@ static int tegra_mc_probe(struct platform_device *pdev)
> >  
> >  	mc->debugfs.root = debugfs_create_dir("mc", NULL);
> >  
> > -	if (mc->soc->init) {
> > -		err = mc->soc->init(mc);
> > +	if (mc->soc->ops && mc->soc->ops->init) {
> > +		err = mc->soc->ops->init(mc);
> >  		if (err < 0)
> >  			dev_err(&pdev->dev, "failed to initialize SoC driver: %d\n",
> >  				err);
> > @@ -867,30 +867,22 @@ static int tegra_mc_probe(struct platform_device *pdev)
> >  	return 0;
> >  }
> >  
> > -static int tegra_mc_suspend(struct device *dev)
> > +static int __maybe_unused tegra_mc_suspend(struct device *dev)
> >  {
> >  	struct tegra_mc *mc = dev_get_drvdata(dev);
> > -	int err;
> >  
> > -	if (IS_ENABLED(CONFIG_TEGRA_IOMMU_GART) && mc->gart) {
> > -		err = tegra_gart_suspend(mc->gart);
> > -		if (err)
> > -			return err;
> > -	}
> > +	if (mc->soc->ops && mc->soc->ops->suspend)
> > +		return mc->soc->ops->suspend(mc);
> >  
> >  	return 0;
> >  }
> >  
> > -static int tegra_mc_resume(struct device *dev)
> > +static int __maybe_unused tegra_mc_resume(struct device *dev)
> >  {
> >  	struct tegra_mc *mc = dev_get_drvdata(dev);
> > -	int err;
> >  
> > -	if (IS_ENABLED(CONFIG_TEGRA_IOMMU_GART) && mc->gart) {
> > -		err = tegra_gart_resume(mc->gart);
> > -		if (err)
> > -			return err;
> > -	}
> > +	if (mc->soc->ops && mc->soc->ops->resume)
> > +		return mc->soc->ops->resume(mc);
> >  
> >  	return 0;
> >  }
> > diff --git a/drivers/memory/tegra/tegra186.c b/drivers/memory/tegra/tegra186.c
> > index 8e77567d1378..9d3fdb609d55 100644
> > --- a/drivers/memory/tegra/tegra186.c
> > +++ b/drivers/memory/tegra/tegra186.c
> > @@ -45,6 +45,17 @@ static void tegra186_mc_program_sid(struct tegra_mc *mc)
> >  	}
> >  }
> >  
> > +static int tegra186_mc_resume(struct tegra_mc *mc)
> > +{
> > +	tegra186_mc_program_sid(mc);
> > +
> > +	return 0;
> > +}
> > +
> > +static const struct tegra_mc_ops tegra186_mc_ops = {
> > +	.resume = tegra186_mc_resume,
> > +};
> > +
> >  #if defined(CONFIG_ARCH_TEGRA_186_SOC)
> >  static const struct tegra_mc_client tegra186_mc_clients[] = {
> >  	{
> > @@ -701,6 +712,7 @@ static const struct tegra_mc_client tegra186_mc_clients[] = {
> >  static const struct tegra_mc_soc tegra186_mc_soc = {
> >  	.num_clients = ARRAY_SIZE(tegra186_mc_clients),
> >  	.clients = tegra186_mc_clients,
> > +	.ops = &tegra186_mc_ops,
> >  };
> >  #endif
> >  
> > @@ -1909,6 +1921,7 @@ static const struct tegra_mc_client tegra194_mc_clients[] = {
> >  static const struct tegra_mc_soc tegra194_mc_soc = {
> >  	.num_clients = ARRAY_SIZE(tegra194_mc_clients),
> >  	.clients = tegra194_mc_clients,
> > +	.ops = &tegra186_mc_ops,
> >  };
> >  #endif
> >  
> > @@ -1961,22 +1974,28 @@ static const struct of_device_id tegra186_mc_of_match[] = {
> >  };
> >  MODULE_DEVICE_TABLE(of, tegra186_mc_of_match);
> >  
> > -static int __maybe_unused tegra186_mc_suspend(struct device *dev)
> > +static int __maybe_unused tegra_mc_suspend(struct device *dev)
> >  {
> > +	struct tegra_mc *mc = dev_get_drvdata(dev);
> > +
> > +	if (mc->soc->ops && mc->soc->ops->suspend)
> > +		return mc->soc->ops->suspend(mc);
> > +
> >  	return 0;
> >  }
> >  
> > -static int __maybe_unused tegra186_mc_resume(struct device *dev)
> > +static int __maybe_unused tegra_mc_resume(struct device *dev)
> >  {
> >  	struct tegra_mc *mc = dev_get_drvdata(dev);
> >  
> > -	tegra186_mc_program_sid(mc);
> > +	if (mc->soc->ops && mc->soc->ops->resume)
> > +		return mc->soc->ops->resume(mc);
> >  
> >  	return 0;
> >  }
> >  
> >  static const struct dev_pm_ops tegra186_mc_pm_ops = {
> > -	SET_SYSTEM_SLEEP_PM_OPS(tegra186_mc_suspend, tegra186_mc_resume)
> > +	SET_SYSTEM_SLEEP_PM_OPS(tegra_mc_suspend, tegra_mc_resume)
> 
> What's the benefit here? You basically define your own suspend-resume
> ops, on top of PM suspend-resume ops... Before it was quite obvious code
> - the Tegra186 MC driver had very simple suspend/resume which did simple
> job. Now it feels like trickier code to follow - Tegra186 driver calls
> it's resume (with the same name as others - another confusion) which is
> a simple wrapper calling somewhere else (need to jump to assinment of
> resume()).

This confusion is merely temporary. The idea here is to make the
Tegra186 driver the same as the "legacy" driver in this regard so that
the unification patch becomes just a removal of this code. If I didn't
first make this the same, I would've had to push this set of changes
into the unification patch, making it more difficult to understand.

Thierry

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[-- Attachment #2: Type: text/plain, Size: 176 bytes --]

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  reply	other threads:[~2021-04-26 12:25 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-20 16:52 [PATCH 00/10] memory: tegra: Driver unification Thierry Reding
2021-04-20 16:52 ` [PATCH 02/10] memory: tegra: Unify struct tegra_mc across SoC generations Thierry Reding
2021-04-20 16:52 ` [PATCH 03/10] memory: tegra: Push suspend/resume into SoC drivers Thierry Reding
2021-04-22  0:06   ` Dmitry Osipenko
2021-04-26  8:47   ` Krzysztof Kozlowski
2021-04-26 12:24     ` Thierry Reding [this message]
2021-04-20 16:52 ` [PATCH 04/10] memory: tegra: Make per-SoC setup more generic Thierry Reding
2021-04-26  8:49   ` Krzysztof Kozlowski
2021-04-26 12:21     ` Thierry Reding
2021-04-20 16:52 ` [PATCH 05/10] memory: tegra: Extract setup code into callback Thierry Reding
2021-04-26  8:58   ` Krzysztof Kozlowski
2021-04-26 12:20     ` Thierry Reding
2021-04-20 16:52 ` [PATCH 06/10] memory: tegra: Parameterize interrupt handler Thierry Reding
2021-04-21 23:33   ` Dmitry Osipenko
2021-04-26  9:01   ` Krzysztof Kozlowski
2021-04-26 12:19     ` Thierry Reding
2021-04-20 16:52 ` [PATCH 07/10] memory: tegra: Only initialize reset controller if available Thierry Reding
2021-04-20 16:52 ` [PATCH 08/10] memory: tegra: Unify drivers Thierry Reding
2021-04-21 11:46   ` Dmitry Osipenko
2021-04-26 12:15     ` Thierry Reding
2021-04-20 16:52 ` [PATCH 09/10] memory: tegra: Add memory client IDs to tables Thierry Reding
2021-04-26  9:06 ` [PATCH 00/10] memory: tegra: Driver unification Krzysztof Kozlowski
2021-04-26 12:15   ` Thierry Reding

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=YIaxDk39hjMJ6nxA@orome.fritz.box \
    --to=thierry.reding@gmail.com \
    --cc=digetx@gmail.com \
    --cc=jonathanh@nvidia.com \
    --cc=krzysztof.kozlowski@canonical.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-tegra@vger.kernel.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;
as well as URLs for NNTP newsgroup(s).