linux-kernel - Re: [PATCH] mfd: qcom_rpm: handle message RAM clock

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20160818141701.GS26240@tuxbot>
Date:	Thu, 18 Aug 2016 07:17:01 -0700
From:	Bjorn Andersson <bjorn.andersson@...aro.org>
To:	Linus Walleij <linus.walleij@...aro.org>
Cc:	lee.jones@...aro.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mfd: qcom_rpm: handle message RAM clock

On Thu 18 Aug 06:00 PDT 2016, Linus Walleij wrote:

> The MSM8660, APQ8060, IPQ806x and MSM8960 have a GCC clock
> to the message RAM used by the RPM. This needs to be enabled
> for messages to pass through. This is a crude solution that
> simply prepare/enable at probe() and disable/unprepare
> at remove(). More elaborate PM is probably possible to
> add later.
> 
> The construction uses IS_ERR() to gracefully handle the
> platforms that do not provide a message RAM clock. It will
> bail out of probe only if the clock is hitting a probe
> deferral situation.
> 

Even more convenient is that clk_prepare_enable() and
clk_disable_unprepare() accepts a NULL.

So set rpm->ramclk = NULL in the error path and move the
clk_enable/disable() out of their conditional.

> Of course this requires the proper device tree set-up:
> 
> rpm: rpm@...000 {
>     compatible = "qcom,rpm-msm8660";
>     clocks = <&gcc RPM_MSG_RAM_H_CLK>;
>     clock-names = "ram";
>     ...
> };
> 

Please write a few lines in the binding document as well:
Documentation/devicetree/bindings/mfd/qcom-rpm.txt

> I have provided this in the MSM8660 device tree, and will
> provide patches for the other targets.
> 
> Cc: Björn Andersson <bjorn.andersson@...aro.org>
> Signed-off-by: Linus Walleij <linus.walleij@...aro.org>
> ---
>  drivers/mfd/qcom_rpm.c | 20 ++++++++++++++++++++
>  1 file changed, 20 insertions(+)
> 
> diff --git a/drivers/mfd/qcom_rpm.c b/drivers/mfd/qcom_rpm.c
[..]
> @@ -501,6 +503,20 @@ static int qcom_rpm_probe(struct platform_device *pdev)
>  	mutex_init(&rpm->lock);
>  	init_completion(&rpm->ack);
>  
> +	/* Enable message RAM clock */
> +	rpm->ramclk = devm_clk_get(&pdev->dev, "ram");
> +	if (IS_ERR(rpm->ramclk)) {
> +		ret = PTR_ERR(rpm->ramclk);
> +		if (ret == -EPROBE_DEFER)
> +			return ret;
> +		/*
> +		 * Fall through in all other cases, as the clock is
> +		 * optional. (Does not exist on all platforms.)
> +		 */

rpm->ramclk = NULL;

> +	} else {
> +		clk_prepare_enable(rpm->ramclk);
> +	}
> +
>  	irq_ack = platform_get_irq_byname(pdev, "ack");
>  	if (irq_ack < 0) {
>  		dev_err(&pdev->dev, "required ack interrupt missing\n");
> @@ -620,6 +636,10 @@ static int qcom_rpm_probe(struct platform_device *pdev)
>  
>  static int qcom_rpm_remove(struct platform_device *pdev)
>  {
> +	struct qcom_rpm *rpm = dev_get_drvdata(&pdev->dev);
> +
> +	if (!IS_ERR(rpm->ramclk))
> +		clk_disable_unprepare(rpm->ramclk);

I suggest that you do this after of_platform_depopulate().

>  	of_platform_depopulate(&pdev->dev);
>  	return 0;
>  }

Regards,
Bjorn