From: Claudiu Beznea <claudiu.beznea@kernel.org>
To: Aubin Constans <aubin.constans@microchip.com>,
Ulf Hansson <ulfh@kernel.org>,
Adrian Hunter <adrian.hunter@intel.com>,
Eugen Hristev <ehristev@kernel.org>,
Romain Sioen <romain.sioen@microchip.com>,
Nicolas Ferre <nicolas.ferre@microchip.com>,
Ludovic Desroches <ludovic.desroches@microchip.com>,
Robert Marko <robert.marko@sartura.hr>
Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>,
Ryan Wanner <ryan.wanner@microchip.com>,
linux-mmc@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] mmc: sdhci-of-at91: disable selectively clocks on pm runtime
Date: Wed, 23 Sep 2026 11:41:57 +0300 [thread overview]
Message-ID: <ead48f50-5d37-406c-b51e-d754c2c512f8@kernel.org> (raw)
In-Reply-To: <20260918153414.909333-1-aubin.constans@microchip.com>
Hi, Aubin,
On 9/18/26 18:34, Aubin Constans wrote:
> From: Claudiu Beznea <claudiu.beznea@tuxon.dev>
>
> SAMA7G5's SDMMC run-time clock disabling is not supported.
> Add support to avoid this scenario for SAMA7G5.
>
> [Amended with a fix from Romain Sioen below this line]
This should go close to the Romain's SoB.
>
> Sama7 boards are impacted by an unbalanced prepare/unprepare
> and enable/disable clock management in sdhci driver due to
> the lack of support for SDMMC clock runtime disabling,
> managed instead by the PMC driver.
>
> Add conditions in the preset function to manage clock
> depending on if the board support runtime clock disabling
> and if we come from the probing or restoring procedure.
>
> Add boolean start_clks variable to manage condition
> in preset function.
>
> Signed-off-by: Claudiu Beznea <claudiu.beznea@tuxon.dev>
Maybe better use the @microchip.com emails, even though that is not available
anymore, to give Microchip credit for that and avoid any legal issues, if any.
There is a mailmap for claudiu.beznea@microchip.com available in the .mailmap
file, if any.
Thank you,
Claudiu
> Co-developed-by: Romain Sioen <romain.sioen@microchip.com>
> Signed-off-by: Romain Sioen <romain.sioen@microchip.com>
> Signed-off-by: Aubin Constans <aubin.constans@microchip.com>
> ---
> drivers/mmc/host/sdhci-of-at91.c | 36 ++++++++++++++++++++++++--------
> 1 file changed, 27 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/mmc/host/sdhci-of-at91.c b/drivers/mmc/host/sdhci-of-at91.c
> index 7c4ac65f247d..399856643e57 100644
> --- a/drivers/mmc/host/sdhci-of-at91.c
> +++ b/drivers/mmc/host/sdhci-of-at91.c
> @@ -39,6 +39,7 @@ struct sdhci_at91_soc_data {
> const struct sdhci_pltfm_data *pdata;
> bool baseclk_is_generated_internally;
> unsigned int divider_for_baseclk;
> + bool pm_runtime_disable_clks;
> };
>
> struct sdhci_at91_priv {
> @@ -149,12 +150,14 @@ static const struct sdhci_pltfm_data sdhci_sama5d2_pdata = {
> static const struct sdhci_at91_soc_data soc_data_sama5d2 = {
> .pdata = &sdhci_sama5d2_pdata,
> .baseclk_is_generated_internally = false,
> + .pm_runtime_disable_clks = true,
> };
>
> static const struct sdhci_at91_soc_data soc_data_sam9x60 = {
> .pdata = &sdhci_sama5d2_pdata,
> .baseclk_is_generated_internally = true,
> .divider_for_baseclk = 2,
> + .pm_runtime_disable_clks = true,
> };
>
> static const struct of_device_id sdhci_at91_dt_match[] = {
> @@ -164,7 +167,7 @@ static const struct of_device_id sdhci_at91_dt_match[] = {
> };
> MODULE_DEVICE_TABLE(of, sdhci_at91_dt_match);
>
> -static int sdhci_at91_set_clks_presets(struct device *dev)
> +static int sdhci_at91_set_clks_presets(struct device *dev, bool start_clks)
> {
> struct sdhci_host *host = dev_get_drvdata(dev);
> struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
> @@ -174,7 +177,14 @@ static int sdhci_at91_set_clks_presets(struct device *dev)
> unsigned int gck_rate, clk_base_rate;
> unsigned int preset_div;
>
> - clk_prepare_enable(priv->hclock);
> + /*
> + * Manage clock prepare/enable procedure depending on if the board
> + * support runtime clock disabling and if we come from the probing
> + * or restoring procedure (for backup+self refresh).
> + */
> + if (start_clks)
> + clk_prepare_enable(priv->hclock);
> +
> caps0 = readl(host->ioaddr + SDHCI_CAPABILITIES);
> caps1 = readl(host->ioaddr + SDHCI_CAPABILITIES_1);
>
> @@ -223,8 +233,11 @@ static int sdhci_at91_set_clks_presets(struct device *dev)
> writew(SDHCI_AT91_PRESET_COMMON_CONF | preset_div,
> host->ioaddr + SDHCI_PRESET_FOR_DDR50);
>
> - clk_prepare_enable(priv->mainck);
> - clk_prepare_enable(priv->gck);
> + /* Same clock management as for hclock */
> + if (start_clks) {
> + clk_prepare_enable(priv->mainck);
> + clk_prepare_enable(priv->gck);
> + }
>
> return 0;
> }
> @@ -254,9 +267,11 @@ static int sdhci_at91_runtime_suspend(struct device *dev)
> if (host->tuning_mode != SDHCI_TUNING_MODE_3)
> mmc_retune_needed(host->mmc);
>
> - clk_disable_unprepare(priv->gck);
> - clk_disable_unprepare(priv->hclock);
> - clk_disable_unprepare(priv->mainck);
> + if (priv->soc_data->pm_runtime_disable_clks) {
> + clk_disable_unprepare(priv->gck);
> + clk_disable_unprepare(priv->hclock);
> + clk_disable_unprepare(priv->mainck);
> + }
>
> return 0;
> }
> @@ -269,7 +284,7 @@ static int sdhci_at91_runtime_resume(struct device *dev)
> int ret;
>
> if (priv->restore_needed) {
> - ret = sdhci_at91_set_clks_presets(dev);
> + ret = sdhci_at91_set_clks_presets(dev, priv->soc_data->pm_runtime_disable_clks);
> if (ret)
> return ret;
>
> @@ -277,6 +292,9 @@ static int sdhci_at91_runtime_resume(struct device *dev)
> goto out;
> }
>
> + if (!priv->soc_data->pm_runtime_disable_clks)
> + goto out;
> +
> ret = clk_prepare_enable(priv->mainck);
> if (ret) {
> dev_err(dev, "can't enable mainck\n");
> @@ -344,7 +362,7 @@ static int sdhci_at91_probe(struct platform_device *pdev)
> return dev_err_probe(&pdev->dev, PTR_ERR(priv->gck),
> "failed to get multclk\n");
>
> - ret = sdhci_at91_set_clks_presets(&pdev->dev);
> + ret = sdhci_at91_set_clks_presets(&pdev->dev, true);
> if (ret)
> return ret;
>
next prev parent reply other threads:[~2026-09-23 8:42 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-18 15:34 Aubin Constans
2026-09-22 12:04 ` Robert Marko
2026-09-23 8:41 ` Claudiu Beznea [this message]
2026-09-23 14:35 ` Adrian Hunter
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=ead48f50-5d37-406c-b51e-d754c2c512f8@kernel.org \
--to=claudiu.beznea@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=alexandre.belloni@bootlin.com \
--cc=aubin.constans@microchip.com \
--cc=ehristev@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mmc@vger.kernel.org \
--cc=ludovic.desroches@microchip.com \
--cc=nicolas.ferre@microchip.com \
--cc=robert.marko@sartura.hr \
--cc=romain.sioen@microchip.com \
--cc=ryan.wanner@microchip.com \
--cc=ulfh@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
all inboxes | Powered by JetHome®