[PATCH v3 0/1] ahci: imx: setup power saving methods

[PATCH v3 0/1] ahci: imx: setup power saving methods

[Richard Zhu]

v3: ahci: imx: setup power saving methods
 - Thanks for Tejun's suggestions.
 override the error_handler by ahci_imx_error_handler, to do
 the complete SATA device detection, enter into PDDQ mode,
 if there is no any SATA device.

 - minor modifications:
   - The format of the copyright is changed, because that the
   original one can't pass fsl internal patch reivew without
   the character '(c)'.
   - Change the module parameters to be more readable.

V2: http://www.spinics.net/lists/linux-ide/msg46606.html
 - kernel build option is replaced by the module parameter(ahci-imx.hp) 

v1: http://www.spinics.net/lists/linux-ide/msg46389.html
 * Disable sata phy internal pll reference clock when
 sysetem enter into suspend mode, enable it after resume.
 * Enter into test power down mode when there is no sata disk
 detected on the port and 'AHCI_IMX_PHY_POWER_DOWN_MODE' is
 enabled.

This patch is based on for-next branch of
"http://git.kernel.org/pub/scm/linux/kernel/git/tj/libata.git"


[v3] ahci: imx: setup power saving methods
drivers/ata/ahci_imx.c |  125 +++++++++++++++++++++++++++++++++++++++++++++++-
1 files changed, 123 insertions(+), 2 deletions(-)

[v3] ahci: imx: setup power saving methods

[Tejun Heo]

Hello,

On Fri, Oct 11, 2013 at 06:37:11PM +0800, Richard Zhu wrote:
> @@ -1,6 +1,6 @@
>  /*
> + * copyright (c) 2013 Freescale Semiconductor, Inc.
>   * Freescale IMX AHCI SATA platform driver
> - * Copyright 2013 Freescale Semiconductor, Inc.

Forgot to mention the above in the commit message?

> +static void ahci_imx_error_handler(struct ata_port *ap);
> +static void ahci_host_stop(struct ata_host *host);

Please put the ops definition below the function defs so that the
above aren't necessary.

> +static void ahci_imx_error_handler(struct ata_port *ap)
> +{
> +	u32 reg_val;
> +	struct ata_device *dev;
> +	struct ata_host *host = dev_get_drvdata(ap->dev);
> +	struct ahci_host_priv *hpriv = host->private_data;
> +	void __iomem *mmio = hpriv->mmio;
> +	struct imx_ahci_priv *imxpriv = dev_get_drvdata(ap->dev->parent);
> +
> +	/* wrapper of the ahci_error_handler */
> +	if (!(ap->pflags & ATA_PFLAG_FROZEN)) {
> +		/* restart engine */
> +		ahci_stop_engine(ap);
> +		ahci_start_engine(ap);
> +	}

Please export ahci_error_handler() and use that instead of duplicating
this part.

> +
> +	sata_pmp_error_handler(ap);
> +
> +	if (!ata_dev_enabled(ap->link.device))
> +		ahci_stop_engine(ap);
> +	/* end of wrapper */
> +
> +	if (!(imxpriv->first_time) || ahci_imx_hotplug)
> +		return;
> +
> +	imxpriv->first_time = false;
> +
> +	ata_for_each_dev(dev, &ap->link, ENABLED)
> +		return;
> +	/* DISABLE LINK to save power consumption */

A bit more explanation would be nice.  ie. briefly explain why it
can't be part of usual LPM.

> +	reg_val = readl(mmio + PORT_PHY_CTL);
> +	writel(reg_val | PORT_PHY_CTL_PDDQ_LOC, mmio + PORT_PHY_CTL);
> +	regmap_update_bits(imxpriv->gpr, IOMUXC_GPR13,
> +			IMX6Q_GPR13_SATA_MPLL_CLK_EN,
> +			!IMX6Q_GPR13_SATA_MPLL_CLK_EN);
> +	clk_disable_unprepare(imxpriv->sata_ref_clk);
> +	imxpriv->no_device = true;
> +}
> +
> +static void ahci_host_stop(struct ata_host *host)
> +{
> +	struct device *dev = host->dev;
> +	struct ahci_platform_data *pdata = dev_get_platdata(dev);
> +	struct ahci_host_priv *hpriv = host->private_data;
> +
> +	if (pdata && pdata->exit)
> +		pdata->exit(dev);
> +
> +	if (!IS_ERR(hpriv->clk)) {
> +		clk_disable_unprepare(hpriv->clk);
> +		clk_put(hpriv->clk);
> +	}
> +}

Please do not duplicate functions like this.  Why not export and
inherit from ahci_platform_ops?

>  static int imx6q_sata_init(struct device *dev, void __iomem *mmio)
>  {
>  	int ret = 0;
> -	unsigned int reg_val;
> +	u32 reg_val;

Hah?  Why is this chunk in this patch?

> +static int imx_ahci_suspend(struct device *dev)
> +{
> +	struct imx_ahci_priv *imxpriv =  dev_get_drvdata(dev->parent);
> +
> +	if (!(imxpriv->no_device)) {

Please lose the unnecessary parantheses.

> +		regmap_update_bits(imxpriv->gpr, IOMUXC_GPR13,
> +				IMX6Q_GPR13_SATA_MPLL_CLK_EN,
> +				!IMX6Q_GPR13_SATA_MPLL_CLK_EN);
> +		clk_disable_unprepare(imxpriv->sata_ref_clk);
> +	}

Some comment of what's going on would be nice.

Other than the above, looks good to me.

Thanks.

-- 
tejun

[v3] ahci: imx: setup power saving methods

[Zhu Richard-R65037]

Hi Tejun:
Thanks for your review comments.

Best Regards
Richard Zhu


-----Original Message-----
From: Tejun Heo [mailto:htejun at gmail.com] On Behalf Of Tejun Heo
Sent: Monday, October 14, 2013 4:17 AM
To: Richard Zhu
Cc: shawn.guo at linaro.org; linux-arm-kernel at lists.infradead.org; jgarzik at pobox.com; linux-ide at vger.kernel.org; Zhu Richard-R65037
Subject: Re: [v3] ahci: imx: setup power saving methods

Hello,

On Fri, Oct 11, 2013 at 06:37:11PM +0800, Richard Zhu wrote:
> @@ -1,6 +1,6 @@
>  /*
> + * copyright (c) 2013 Freescale Semiconductor, Inc.
>   * Freescale IMX AHCI SATA platform driver
> - * Copyright 2013 Freescale Semiconductor, Inc.

Forgot to mention the above in the commit message?
[Richard] Ok, I would add change-note in the commit message.

> +static void ahci_imx_error_handler(struct ata_port *ap); static void 
> +ahci_host_stop(struct ata_host *host);

Please put the ops definition below the function defs so that the above aren't necessary.
[Richard] Accept.

> +static void ahci_imx_error_handler(struct ata_port *ap) {
> +	u32 reg_val;
> +	struct ata_device *dev;
> +	struct ata_host *host = dev_get_drvdata(ap->dev);
> +	struct ahci_host_priv *hpriv = host->private_data;
> +	void __iomem *mmio = hpriv->mmio;
> +	struct imx_ahci_priv *imxpriv = dev_get_drvdata(ap->dev->parent);
> +
> +	/* wrapper of the ahci_error_handler */
> +	if (!(ap->pflags & ATA_PFLAG_FROZEN)) {
> +		/* restart engine */
> +		ahci_stop_engine(ap);
> +		ahci_start_engine(ap);
> +	}

Please export ahci_error_handler() and use that instead of duplicating this part.
[Richard] Ok, accept.

> +
> +	sata_pmp_error_handler(ap);
> +
> +	if (!ata_dev_enabled(ap->link.device))
> +		ahci_stop_engine(ap);
> +	/* end of wrapper */
> +
> +	if (!(imxpriv->first_time) || ahci_imx_hotplug)
> +		return;
> +
> +	imxpriv->first_time = false;
> +
> +	ata_for_each_dev(dev, &ap->link, ENABLED)
> +		return;
> +	/* DISABLE LINK to save power consumption */

A bit more explanation would be nice.  ie. briefly explain why it can't be part of usual LPM.
[Richard] no problem, I would add the explanation later. Thanks.

> +	reg_val = readl(mmio + PORT_PHY_CTL);
> +	writel(reg_val | PORT_PHY_CTL_PDDQ_LOC, mmio + PORT_PHY_CTL);
> +	regmap_update_bits(imxpriv->gpr, IOMUXC_GPR13,
> +			IMX6Q_GPR13_SATA_MPLL_CLK_EN,
> +			!IMX6Q_GPR13_SATA_MPLL_CLK_EN);
> +	clk_disable_unprepare(imxpriv->sata_ref_clk);
> +	imxpriv->no_device = true;
> +}
> +
> +static void ahci_host_stop(struct ata_host *host) {
> +	struct device *dev = host->dev;
> +	struct ahci_platform_data *pdata = dev_get_platdata(dev);
> +	struct ahci_host_priv *hpriv = host->private_data;
> +
> +	if (pdata && pdata->exit)
> +		pdata->exit(dev);
> +
> +	if (!IS_ERR(hpriv->clk)) {
> +		clk_disable_unprepare(hpriv->clk);
> +		clk_put(hpriv->clk);
> +	}
> +}

Please do not duplicate functions like this.  Why not export and inherit from ahci_platform_ops?
[Richard] Ok, I would export and inherit from ahci_platform_ops.

>  static int imx6q_sata_init(struct device *dev, void __iomem *mmio)  {
>  	int ret = 0;
> -	unsigned int reg_val;
> +	u32 reg_val;

Hah?  Why is this chunk in this patch?
[Richard] Would be removed later. Thanks.

> +static int imx_ahci_suspend(struct device *dev) {
> +	struct imx_ahci_priv *imxpriv =  dev_get_drvdata(dev->parent);
> +
> +	if (!(imxpriv->no_device)) {

Please lose the unnecessary parantheses.
[Richard] Would be removed later. Thanks.

> +		regmap_update_bits(imxpriv->gpr, IOMUXC_GPR13,
> +				IMX6Q_GPR13_SATA_MPLL_CLK_EN,
> +				!IMX6Q_GPR13_SATA_MPLL_CLK_EN);
> +		clk_disable_unprepare(imxpriv->sata_ref_clk);
> +	}

Some comment of what's going on would be nice.

Other than the above, looks good to me.
[Richard] Ok, no problem, comments would be added in next version.
Thanks again. :).

Thanks.

--
tejun