[PATCH 16/18] PCI: aardvark: Add suspend to RAM support

[Lorenzo Pieralisi]

[+Rafael]

On Sun, Feb 20, 2022 at 08:33:44PM +0100, Marek Behún wrote:
> From: Miquel Raynal <miquel.raynal at bootlin.com>
> 
> Add suspend and resume callbacks. The priority of these are
> "_noirq()", to workaround early access to the registers done by the
> PCI core through the ->read()/->write() callbacks at resume time.

This commit log should be rewritten to clarify it. IIUC, you are hooking
up the _noirq() callbacks to take advantage of callbacks ordering rather
than needing the IRQ disabled per-se.

Thread here:
https://lore.kernel.org/linux-pci/20220220193346.23789-17-kabel@kernel.org

I'd ask Rafael please if this is common practice or a workaround and
whether that's how it should be done.

Thanks,
Lorenzo

> Signed-off-by: Miquel Raynal <miquel.raynal at bootlin.com>
> Signed-off-by: Pali Rohár <pali at kernel.org>
> Signed-off-by: Marek Behún <kabel at kernel.org>
> ---
>  drivers/pci/controller/pci-aardvark.c | 39 +++++++++++++++++++++++++++
>  1 file changed, 39 insertions(+)
> 
> diff --git a/drivers/pci/controller/pci-aardvark.c b/drivers/pci/controller/pci-aardvark.c
> index 3b51f47abd72..8c9ac7766ac7 100644
> --- a/drivers/pci/controller/pci-aardvark.c
> +++ b/drivers/pci/controller/pci-aardvark.c
> @@ -1896,6 +1896,44 @@ static int advk_pcie_setup_phy(struct advk_pcie *pcie)
>  	return ret;
>  }
>  
> +static int __maybe_unused advk_pcie_suspend(struct device *dev)
> +{
> +	struct advk_pcie *pcie = dev_get_drvdata(dev);
> +
> +	advk_pcie_disable_phy(pcie);
> +
> +	clk_disable_unprepare(pcie->clk);
> +
> +	return 0;
> +}
> +
> +static int __maybe_unused advk_pcie_resume(struct device *dev)
> +{
> +	struct advk_pcie *pcie = dev_get_drvdata(dev);
> +	int ret;
> +
> +	ret = clk_prepare_enable(pcie->clk);
> +	if (ret)
> +		return ret;
> +
> +	ret = advk_pcie_enable_phy(pcie);
> +	if (ret)
> +		return ret;
> +
> +	advk_pcie_setup_hw(pcie);
> +
> +	return 0;
> +}
> +
> +/*
> + * The PCI core will try to reconfigure the bus quite early in the resume path.
> + * We must use the _noirq() alternatives to ensure the controller is ready when
> + * the core uses the ->read()/->write() callbacks.
> + */
> +static const struct dev_pm_ops advk_pcie_dev_pm_ops = {
> +	SET_NOIRQ_SYSTEM_SLEEP_PM_OPS(advk_pcie_suspend, advk_pcie_resume)
> +};
> +
>  static int advk_pcie_probe(struct platform_device *pdev)
>  {
>  	struct device *dev = &pdev->dev;
> @@ -2171,6 +2209,7 @@ static struct platform_driver advk_pcie_driver = {
>  	.driver = {
>  		.name = "advk-pcie",
>  		.of_match_table = advk_pcie_of_match_table,
> +		.pm = &advk_pcie_dev_pm_ops,
>  	},
>  	.probe = advk_pcie_probe,
>  	.remove = advk_pcie_remove,
> -- 
> 2.34.1
>