linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Enric Balletbo Serra <eballetbo@gmail.com>
To: Eizan Miyamoto <eizan@chromium.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Andrew-CT Chen <andrew-ct.chen@mediatek.com>,
	Minghsiu Tsai <minghsiu.tsai@mediatek.com>,
	Houlong Wei <houlong.wei@mediatek.com>,
	"moderated list:ARM/Mediatek SoC support" 
	<linux-mediatek@lists.infradead.org>,
	Matthias Brugger <matthias.bgg@gmail.com>,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	Linux ARM <linux-arm-kernel@lists.infradead.org>,
	linux-media@vger.kernel.org
Subject: Re: [PATCH v2 2/2] [media] mtk-mdp: use pm_runtime in MDP component driver
Date: Wed, 6 May 2020 19:06:52 +0200	[thread overview]
Message-ID: <CAFqH_50_Py7Diu5bwmjLBPGiy3F5J5qNGmtrbH7i7aTUaHif9A@mail.gmail.com> (raw)
In-Reply-To: <20200506184018.v2.2.I87cf35a058328c780bd3b8b2191209a5011b7016@changeid>

Hi Eizan,

Thank you for the patch.

Missatge de Eizan Miyamoto <eizan@chromium.org> del dia dc., 6 de maig
2020 a les 10:42:
>
> Without this change, the MDP components are not fully integrated into
> the runtime power management subsystem, and the MDP driver does not
> work.
>
> For each of the component device drivers to be able to call
> pm_runtime_get/put_sync() a pointer to the component's device struct
> had to be added to struct mtk_mdp_comp, set by mtk_mdp_comp_init().
>
> Note that the dev argument to mtk_mdp_comp_clock_on/off() has been
> removed. Those functions used to be called from the "master" mdp driver
> in mtk_mdp_core.c, but the component's device pointer no longer
> corresponds to the mdp master device pointer, which is not the right
> device to pass to pm_runtime_put/get_sync() which we had to add to get
> the driver to work properly.
>
> Signed-off-by: Eizan Miyamoto <eizan@chromium.org>
> ---
>
> Changes in v2:

Ah, I guess this change log corresponds to the first patch.

> - remove empty mtk_mdp_comp_init
> - update documentation for enum mtk_mdp_comp_type
> - remove comma after last element of mtk_mdp_comp_driver_dt_match
>
>  drivers/media/platform/mtk-mdp/mtk_mdp_comp.c | 22 ++++++++++++++-----
>  drivers/media/platform/mtk-mdp/mtk_mdp_comp.h |  6 +++--
>  drivers/media/platform/mtk-mdp/mtk_mdp_core.c |  6 ++---
>  3 files changed, 23 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/media/platform/mtk-mdp/mtk_mdp_comp.c b/drivers/media/platform/mtk-mdp/mtk_mdp_comp.c
> index 5b4d482df778..228c58f92c8c 100644
> --- a/drivers/media/platform/mtk-mdp/mtk_mdp_comp.c
> +++ b/drivers/media/platform/mtk-mdp/mtk_mdp_comp.c
> @@ -15,6 +15,7 @@
>  #include <linux/of_platform.h>
>  #include <soc/mediatek/smi.h>
>  #include <linux/platform_device.h>
> +#include <linux/pm_runtime.h>
>
>  #include "mtk_mdp_comp.h"
>  #include "mtk_mdp_core.h"
> @@ -53,7 +54,7 @@ static const struct of_device_id mtk_mdp_comp_driver_dt_match[] = {
>  };
>  MODULE_DEVICE_TABLE(of, mtk_mdp_comp_driver_dt_match);
>
> -void mtk_mdp_comp_clock_on(struct device *dev, struct mtk_mdp_comp *comp)
> +void mtk_mdp_comp_clock_on(struct mtk_mdp_comp *comp)
>  {
>         int i, err;
>
> @@ -62,25 +63,31 @@ void mtk_mdp_comp_clock_on(struct device *dev, struct mtk_mdp_comp *comp)
>                 if (err) {
>                         enum mtk_mdp_comp_type comp_type =
>                                 (enum mtk_mdp_comp_type)
> -                               of_device_get_match_data(dev);
> -                       dev_err(dev,
> +                               of_device_get_match_data(comp->dev);
> +                       dev_err(comp->dev,
>                                 "failed to get larb, err %d. type:%d\n",
>                                 err, comp_type);
>                 }
>         }
>
> +       err = pm_runtime_get_sync(comp->dev);
> +       if (err < 0)
> +               dev_err(comp->dev,
> +                       "failed to runtime get, err %d.\n",
> +                       err);

Should you really ignore this error? If that's the case I'd just call
pm_runtime_get_sync() without adding the logic to just print an error
message.

> +
>         for (i = 0; i < ARRAY_SIZE(comp->clk); i++) {
>                 if (IS_ERR(comp->clk[i]))
>                         continue;
>                 err = clk_prepare_enable(comp->clk[i]);
>                 if (err)
> -                       dev_err(dev,
> +                       dev_err(comp->dev,
>                                 "failed to enable clock, err %d. i:%d\n",
>                                 err, i);

Although ignoring errors seems to be a common pattern in this file and
I know you did not introduce this.

>         }
>  }
>
> -void mtk_mdp_comp_clock_off(struct device *dev, struct mtk_mdp_comp *comp)
> +void mtk_mdp_comp_clock_off(struct mtk_mdp_comp *comp)
>  {
>         int i;
>
> @@ -92,6 +99,8 @@ void mtk_mdp_comp_clock_off(struct device *dev, struct mtk_mdp_comp *comp)
>
>         if (comp->larb_dev)
>                 mtk_smi_larb_put(comp->larb_dev);
> +
> +       pm_runtime_put_sync(comp->dev);
>  }
>
>  static int mtk_mdp_comp_bind(struct device *dev, struct device *master,
> @@ -101,6 +110,7 @@ static int mtk_mdp_comp_bind(struct device *dev, struct device *master,
>         struct mtk_mdp_dev *mdp = data;
>
>         mtk_mdp_register_component(mdp, comp);
> +       pm_runtime_enable(dev);
>
>         return 0;
>  }
> @@ -111,6 +121,7 @@ static void mtk_mdp_comp_unbind(struct device *dev, struct device *master,
>         struct mtk_mdp_dev *mdp = data;
>         struct mtk_mdp_comp *comp = dev_get_drvdata(dev);
>
> +       pm_runtime_disable(dev);
>         mtk_mdp_unregister_component(mdp, comp);
>  }
>
> @@ -129,6 +140,7 @@ int mtk_mdp_comp_init(struct mtk_mdp_comp *comp, struct device *dev)
>                  (enum mtk_mdp_comp_type)of_device_get_match_data(dev);
>
>         INIT_LIST_HEAD(&comp->node);
> +       comp->dev = dev;
>
>         for (i = 0; i < ARRAY_SIZE(comp->clk); i++) {
>                 comp->clk[i] = of_clk_get(node, i);
> diff --git a/drivers/media/platform/mtk-mdp/mtk_mdp_comp.h b/drivers/media/platform/mtk-mdp/mtk_mdp_comp.h
> index b5a18fe567aa..de158d3712f6 100644
> --- a/drivers/media/platform/mtk-mdp/mtk_mdp_comp.h
> +++ b/drivers/media/platform/mtk-mdp/mtk_mdp_comp.h
> @@ -12,17 +12,19 @@
>   * @node:      list node to track sibing MDP components
>   * @clk:       clocks required for component
>   * @larb_dev:  SMI device required for component
> + * @dev:       component's device
>   */
>  struct mtk_mdp_comp {
>         struct list_head        node;
>         struct clk              *clk[2];
>         struct device           *larb_dev;
> +       struct device           *dev;
>  };
>
>  int mtk_mdp_comp_init(struct mtk_mdp_comp *comp, struct device *dev);
>
> -void mtk_mdp_comp_clock_on(struct device *dev, struct mtk_mdp_comp *comp);
> -void mtk_mdp_comp_clock_off(struct device *dev, struct mtk_mdp_comp *comp);
> +void mtk_mdp_comp_clock_on(struct mtk_mdp_comp *comp);
> +void mtk_mdp_comp_clock_off(struct mtk_mdp_comp *comp);
>
>  extern struct platform_driver mtk_mdp_component_driver;
>
> diff --git a/drivers/media/platform/mtk-mdp/mtk_mdp_core.c b/drivers/media/platform/mtk-mdp/mtk_mdp_core.c
> index 539a7942e3cb..133d107aa4e6 100644
> --- a/drivers/media/platform/mtk-mdp/mtk_mdp_core.c
> +++ b/drivers/media/platform/mtk-mdp/mtk_mdp_core.c
> @@ -52,20 +52,18 @@ MODULE_DEVICE_TABLE(of, mtk_mdp_of_ids);
>
>  static void mtk_mdp_clock_on(struct mtk_mdp_dev *mdp)
>  {
> -       struct device *dev = &mdp->pdev->dev;
>         struct mtk_mdp_comp *comp_node;
>
>         list_for_each_entry(comp_node, &mdp->comp_list, node)
> -               mtk_mdp_comp_clock_on(dev, comp_node);
> +               mtk_mdp_comp_clock_on(comp_node);
>  }
>
>  static void mtk_mdp_clock_off(struct mtk_mdp_dev *mdp)
>  {
> -       struct device *dev = &mdp->pdev->dev;
>         struct mtk_mdp_comp *comp_node;
>
>         list_for_each_entry(comp_node, &mdp->comp_list, node)
> -               mtk_mdp_comp_clock_off(dev, comp_node);
> +               mtk_mdp_comp_clock_off(comp_node);
>  }
>
>  static void mtk_mdp_wdt_worker(struct work_struct *work)
> --
> 2.26.2.526.g744177e7f7-goog
>
>
> _______________________________________________
> Linux-mediatek mailing list
> Linux-mediatek@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-mediatek

  reply	other threads:[~2020-05-06 17:07 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-06  8:40 [PATCH v2 0/2] Refactor MDP driver and add dummy component driver Eizan Miyamoto
2020-05-06  8:40 ` [PATCH v2 1/2] [media] mtk-mdp: add driver to probe mdp components Eizan Miyamoto
2020-05-06 16:54   ` Enric Balletbo Serra
2020-05-07 11:11     ` Eizan Miyamoto
2020-05-07 13:44       ` Enric Balletbo i Serra
2020-05-15  5:21         ` Eizan Miyamoto
2020-07-20 16:09   ` Chun-Kuang Hu
2020-05-06  8:40 ` [PATCH v2 2/2] [media] mtk-mdp: use pm_runtime in MDP component driver Eizan Miyamoto
2020-05-06 17:06   ` Enric Balletbo Serra [this message]
2020-05-21  1:36     ` Eizan Miyamoto
2020-06-24 14:05 ` [PATCH v2 0/2] Refactor MDP driver and add dummy " Hans Verkuil
2020-06-26  9:37   ` Enric Balletbo i Serra

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=CAFqH_50_Py7Diu5bwmjLBPGiy3F5J5qNGmtrbH7i7aTUaHif9A@mail.gmail.com \
    --to=eballetbo@gmail.com \
    --cc=andrew-ct.chen@mediatek.com \
    --cc=eizan@chromium.org \
    --cc=houlong.wei@mediatek.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-mediatek@lists.infradead.org \
    --cc=matthias.bgg@gmail.com \
    --cc=mchehab@kernel.org \
    --cc=minghsiu.tsai@mediatek.com \
    /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).