From: Adrian Hunter <adrian.hunter@intel.com>
To: Liming Sun <limings@nvidia.com>,
Ulf Hansson <ulf.hansson@linaro.org>,
David Thompson <davthompson@nvidia.com>,
Shawn Lin <shawn.lin@rock-chips.com>
Cc: "linux-mmc@vger.kernel.org" <linux-mmc@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v8] mmc: sdhci-of-dwcmshc: Add runtime PM operations
Date: Fri, 18 Aug 2023 10:27:00 +0300 [thread overview]
Message-ID: <a77d1cc6-e932-6568-fb0f-3afff7fcfb3f@intel.com> (raw)
In-Reply-To: <BN9PR12MB50686B97FCE49645CC849DF6D31AA@BN9PR12MB5068.namprd12.prod.outlook.com>
On 17/08/23 19:22, Liming Sun wrote:
>
>
>> -----Original Message-----
>> From: Adrian Hunter <adrian.hunter@intel.com>
>> Sent: Thursday, August 17, 2023 3:36 AM
>> To: Liming Sun <limings@nvidia.com>; Ulf Hansson <ulf.hansson@linaro.org>;
>> David Thompson <davthompson@nvidia.com>; Shawn Lin <shawn.lin@rock-
>> chips.com>
>> Cc: linux-mmc@vger.kernel.org; linux-kernel@vger.kernel.org
>> Subject: Re: [PATCH v8] mmc: sdhci-of-dwcmshc: Add runtime PM operations
>>
>> On 17/08/23 04:06, Liming Sun wrote:
>>> This commit implements the runtime PM operations to disable eMMC
>>> card clock when idle.
>>>
>>> Reviewed-by: David Thompson <davthompson@nvidia.com>
>>> Signed-off-by: Liming Sun <limings@nvidia.com>
>>> ---
>>> v7->v8:
>>> - Address Ulf's comment (option-1);
>>> - Updates for Adrian;s comment to remove the force_suspend/resume
>>> in dwcmshc_resume()/dwcmshc_suspend(); Add comments for
>>> dwcmshc_resume()/dwcmshc_suspend();
>>> v6->v7:
>>> - Address Ulf's comment;
>>> v5->v6:
>>> - Address Adrian's more comments and add coordination between
>>> runtime PM and system PM;
>>> v4->v5:
>>> - Address Adrian's comment to move the pm_enable to the end to
>>> avoid race;
>>> v3->v4:
>>> - Fix compiling reported by 'kernel test robot';
>>> v2->v3:
>>> - Revise the commit message;
>>> v1->v2:
>>> Updates for comments from Ulf:
>>> - Make the runtime PM logic generic for sdhci-of-dwcmshc;
>>> v1: Initial version.
>>> ---
>>> drivers/mmc/host/sdhci-of-dwcmshc.c | 76
>> +++++++++++++++++++++++++++--
>>> 1 file changed, 73 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/mmc/host/sdhci-of-dwcmshc.c b/drivers/mmc/host/sdhci-
>> of-dwcmshc.c
>>> index e68cd87998c8..2196218c9d79 100644
>>> --- a/drivers/mmc/host/sdhci-of-dwcmshc.c
>>> +++ b/drivers/mmc/host/sdhci-of-dwcmshc.c
>>> @@ -15,6 +15,7 @@
>>> #include <linux/module.h>
>>> #include <linux/of.h>
>>> #include <linux/of_device.h>
>>> +#include <linux/pm_runtime.h>
>>> #include <linux/reset.h>
>>> #include <linux/sizes.h>
>>>
>>> @@ -548,9 +549,13 @@ static int dwcmshc_probe(struct platform_device
>> *pdev)
>>>
>>> host->mmc->caps |= MMC_CAP_WAIT_WHILE_BUSY;
>>>
>>> + pm_runtime_get_noresume(dev);
>>> + pm_runtime_set_active(dev);
>>> + pm_runtime_enable(dev);
>>> +
>>> err = sdhci_setup_host(host);
>>> if (err)
>>> - goto err_clk;
>>> + goto err_rpm;
>>>
>>> if (rk_priv)
>>> dwcmshc_rk35xx_postinit(host, priv);
>>> @@ -559,10 +564,15 @@ static int dwcmshc_probe(struct platform_device
>> *pdev)
>>> if (err)
>>> goto err_setup_host;
>>>
>>> + pm_runtime_put(dev);
>>> +
>>> return 0;
>>>
>>> err_setup_host:
>>> sdhci_cleanup_host(host);
>>> +err_rpm:
>>> + pm_runtime_disable(dev);
>>> + pm_runtime_put_noidle(dev);
>>> err_clk:
>>> clk_disable_unprepare(pltfm_host->clk);
>>> clk_disable_unprepare(priv->bus_clk);
>>> @@ -594,6 +604,10 @@ static int dwcmshc_remove(struct platform_device
>> *pdev)
>>> }
>>>
>>> #ifdef CONFIG_PM_SLEEP
>>> +/*
>>> + * Note, runtime suspend changes only SDHCI_CLOCK_CARD_EN which has
>> no effect on
>>> + * system suspend.
>>> + */
>>
>> This comment isn't needed since pm_runtime_get_sync() will
>> always runtime resume the device if needed.
>
> Updated in v9.
>
>>
>>> static int dwcmshc_suspend(struct device *dev)
>>> {
>>> struct sdhci_host *host = dev_get_drvdata(dev);
>>> @@ -602,9 +616,11 @@ static int dwcmshc_suspend(struct device *dev)
>>> struct rk35xx_priv *rk_priv = priv->priv;
>>> int ret;
>>>
>>> + pm_runtime_get_sync(dev);
>>
>> This needs a corresponding pm_runtime_put() at the end of
>> dwcmshc_resume().
>
> Updated in v9.
>
>>
>>> +
>>> ret = sdhci_suspend_host(host);
>>> if (ret)
>>> - return ret;
>>> + goto err_suspend;
>>>
>>> clk_disable_unprepare(pltfm_host->clk);
>>> if (!IS_ERR(priv->bus_clk))
>>> @@ -614,9 +630,15 @@ static int dwcmshc_suspend(struct device *dev)
>>> clk_bulk_disable_unprepare(RK35xx_MAX_CLKS,
>>> rk_priv->rockchip_clks);
>>>
>>> +err_suspend:
>>> + pm_runtime_put_noidle(dev);
>>> return ret;
>>> }
>>>
>>> +/*
>>> + * Note, system resume leaves SDHCI_CLOCK_INT_EN off which is consistent
>> with
>>> + * either runtime suspended or runtime resumed.
>>> + */
>>
>> As above, this comment isn't needed.
>
> Updated in v9.
>
>>
>>> static int dwcmshc_resume(struct device *dev)
>>
>>
>> As mentioned above, dwcmshc_resume() needs a pm_runtime_put()
>> to balance the pm_runtime_get_sync().
>
> Updated in v9.
>
>>
>> Could fix up the error path too, but that should be a
>> separate prior patch.
>>
>> Probably end up looking like:
>>
>>
>> diff --git a/drivers/mmc/host/sdhci-of-dwcmshc.c b/drivers/mmc/host/sdhci-
>> of-dwcmshc.c
>> index 20aa9b6327d2..c9b020b7a3f6 100644
>> --- a/drivers/mmc/host/sdhci-of-dwcmshc.c
>> +++ b/drivers/mmc/host/sdhci-of-dwcmshc.c
>> @@ -652,17 +652,33 @@ static int dwcmshc_resume(struct device *dev)
>> if (!IS_ERR(priv->bus_clk)) {
>> ret = clk_prepare_enable(priv->bus_clk);
>> if (ret)
>> - return ret;
>> + goto disable_clk;
>> }
>>
>> if (rk_priv) {
>> ret = clk_bulk_prepare_enable(RK35xx_MAX_CLKS,
>> rk_priv->rockchip_clks);
>> if (ret)
>> - return ret;
>> + goto disable_bus_clk;
>> }
>>
>> - return sdhci_resume_host(host);
>> + ret = sdhci_resume_host(host);
>> + if (ret)
>> + goto disable_rockchip_clks;
>> +
>> + pm_runtime_put(dev);
>> +
>> + return 0;
>> +
>> +disable_rockchip_clks:
>> + if (rk_priv)
>> + clk_bulk_disable_unprepare(RK35xx_MAX_CLKS, rk_priv-
>>> rockchip_clks);
>> +disable_bus_clk:
>> + if (!IS_ERR(priv->bus_clk))
>> + clk_disable_unprepare(priv->bus_clk);
>> +disable_clk:
>> + clk_disable_unprepare(pltfm_host->clk);
>> + return ret;
>> }
>> #endif
>>
>>
>
> Do you mean a separate 'post' patch? I left it out in v9 and could post another one for this error path. Or do you prefer the same patch series?
Same patch series, 2 patches. First patch fixes the error paths in dwcmshc_resume().
Second patch adds runtime pm.
>
>>
>>
>>> {
>>> struct sdhci_host *host = dev_get_drvdata(dev);
>>> @@ -646,7 +668,55 @@ static int dwcmshc_resume(struct device *dev)
>>> }
>>> #endif
>>>
>>> -static SIMPLE_DEV_PM_OPS(dwcmshc_pmops, dwcmshc_suspend,
>> dwcmshc_resume);
>>> +#ifdef CONFIG_PM
>>> +
>>> +static void dwcmshc_enable_card_clk(struct sdhci_host *host)
>>> +{
>>> + u16 ctrl;
>>> +
>>> + ctrl = sdhci_readw(host, SDHCI_CLOCK_CONTROL);
>>> + if ((ctrl & SDHCI_CLOCK_INT_EN) && !(ctrl &
>> SDHCI_CLOCK_CARD_EN)) {
>>> + ctrl |= SDHCI_CLOCK_CARD_EN;
>>> + sdhci_writew(host, ctrl, SDHCI_CLOCK_CONTROL);
>>> + }
>>> +}
>>> +
>>> +static void dwcmshc_disable_card_clk(struct sdhci_host *host)
>>> +{
>>> + u16 ctrl;
>>> +
>>> + ctrl = sdhci_readw(host, SDHCI_CLOCK_CONTROL);
>>> + if (ctrl & SDHCI_CLOCK_CARD_EN) {
>>> + ctrl &= ~SDHCI_CLOCK_CARD_EN;
>>> + sdhci_writew(host, ctrl, SDHCI_CLOCK_CONTROL);
>>> + }
>>> +}
>>> +
>>> +static int dwcmshc_runtime_suspend(struct device *dev)
>>> +{
>>> + struct sdhci_host *host = dev_get_drvdata(dev);
>>> +
>>> + dwcmshc_disable_card_clk(host);
>>> +
>>> + return 0;
>>> +}
>>> +
>>> +static int dwcmshc_runtime_resume(struct device *dev)
>>> +{
>>> + struct sdhci_host *host = dev_get_drvdata(dev);
>>> +
>>> + dwcmshc_enable_card_clk(host);
>>> +
>>> + return 0;
>>> +}
>>> +
>>> +#endif
>>> +
>>> +static const struct dev_pm_ops dwcmshc_pmops = {
>>> + SET_SYSTEM_SLEEP_PM_OPS(dwcmshc_suspend, dwcmshc_resume)
>>> + SET_RUNTIME_PM_OPS(dwcmshc_runtime_suspend,
>>> + dwcmshc_runtime_resume, NULL)
>>> +};
>>>
>>> static struct platform_driver sdhci_dwcmshc_driver = {
>>> .driver = {
>
next prev parent reply other threads:[~2023-08-18 7:27 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-11 19:03 [PATCH v1 1/1] mmc: sdhci-of-dwcmshc: Add runtime PM operations for BlueField-3 Liming Sun
2023-05-12 7:35 ` Ulf Hansson
2023-05-12 12:09 ` Liming Sun
2023-05-12 12:20 ` [PATCH v2] " Liming Sun
2023-05-12 12:26 ` [PATCH v3] mmc: sdhci-of-dwcmshc: Add runtime PM operations Liming Sun
2023-05-12 17:43 ` kernel test robot
2023-05-12 18:15 ` [PATCH v4] " Liming Sun
2023-05-19 13:19 ` Adrian Hunter
2023-07-28 12:20 ` Liming Sun
2023-07-28 12:20 ` [PATCH v5] " Liming Sun
2023-08-01 15:36 ` Adrian Hunter
2023-08-04 23:29 ` Liming Sun
2023-08-04 23:30 ` [PATCH v6] " Liming Sun
2023-08-08 9:39 ` Ulf Hansson
2023-08-08 13:21 ` Liming Sun
2023-08-08 13:56 ` Ulf Hansson
2023-08-08 20:24 ` Liming Sun
2023-08-09 10:58 ` Ulf Hansson
2023-08-08 20:23 ` [PATCH v7] " Liming Sun
2023-08-10 8:12 ` Adrian Hunter
2023-08-10 10:21 ` Ulf Hansson
2023-08-10 12:44 ` Adrian Hunter
2023-08-10 16:34 ` Ulf Hansson
2023-08-11 5:56 ` Adrian Hunter
2023-08-11 8:36 ` Ulf Hansson
2023-08-16 16:28 ` Liming Sun
2023-08-16 21:40 ` Ulf Hansson
2023-08-17 1:06 ` [PATCH v8] " Liming Sun
2023-08-17 7:35 ` Adrian Hunter
2023-08-17 16:22 ` Liming Sun
2023-08-18 7:27 ` Adrian Hunter [this message]
2023-08-17 16:21 ` [PATCH v9] " Liming Sun
2023-08-18 9:00 ` Ulf Hansson
2023-08-18 9:35 ` Adrian Hunter
2023-08-18 10:19 ` Ulf Hansson
2023-08-22 19:46 ` Liming Sun
2023-08-22 19:59 ` [PATCH v10 1/2] mmc : sdhci-of-dwcmshc : add error handling in dwcmshc_resume Liming Sun
2023-08-24 7:25 ` Adrian Hunter
2023-08-24 11:00 ` Ulf Hansson
2023-08-22 19:59 ` [PATCH v10 2/2] This commit implements the runtime PM operations to disable eMMC card clock when idle Liming Sun
2023-08-24 7:28 ` Adrian Hunter
2023-08-24 11:00 ` Ulf Hansson
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=a77d1cc6-e932-6568-fb0f-3afff7fcfb3f@intel.com \
--to=adrian.hunter@intel.com \
--cc=davthompson@nvidia.com \
--cc=limings@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mmc@vger.kernel.org \
--cc=shawn.lin@rock-chips.com \
--cc=ulf.hansson@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