Re: [PATCH v12 09/13] PCI: imx6: Turn off regulator when system is in suspend mode

[Bjorn Helgaas]

On Thu, Jun 16, 2022 at 02:23:29AM +0000, Hongxing Zhu wrote:
> > -----Original Message-----
> > From: Bjorn Helgaas <helgaas@xxxxxxxxxx>
> > Sent: 2022年6月16日 7:31
> > To: Hongxing Zhu <hongxing.zhu@xxxxxxx>; Lucas Stach
> > <l.stach@xxxxxxxxxxxxxx>; Rob Herring <robh+dt@xxxxxxxxxx>; Mark Brown
> > <broonie@xxxxxxxxxx>; Lorenzo Pieralisi <lorenzo.pieralisi@xxxxxxx>; Fabio
> > Estevam <festevam@xxxxxxxxx>; Francesco Dolcini
> > <francesco.dolcini@xxxxxxxxxxx>
> > Cc: linux-pci@xxxxxxxxxxxxxxx; linux-arm-kernel@xxxxxxxxxxxxxxxxxxx;
> > linux-kernel@xxxxxxxxxxxxxxx; kernel@xxxxxxxxxxxxxx; dl-linux-imx
> > <linux-imx@xxxxxxx>; Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> > Subject: Re: [PATCH v12 09/13] PCI: imx6: Turn off regulator when system is in
> > suspend mode
> > 
> > On Wed, Jun 15, 2022 at 06:15:47PM -0500, Bjorn Helgaas wrote:
> > > From: Richard Zhu <hongxing.zhu@xxxxxxx>
> > >
> > > The driver should undo any enables it did itself. The regulator
> > > disable shouldn't be basing decisions on regulator_is_enabled().
> > >
> > > Move the regulator_disable to the suspend function, turn off regulator
> > > when the system is in suspend mode.
> > >
> > > To keep the balance of the regulator usage counter, disable the
> > > regulator in shutdown.
> > >
> > > Link:
> > > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore
> > > .kernel.org%2Fr%2F1655189942-12678-6-git-send-email-hongxing.z&amp;d
> > at
> > >
> > a=05%7C01%7Chongxing.zhu%40nxp.com%7C8dea46d908d34cb4825808da4
> > f270eaf%
> > >
> > 7C686ea1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C6379093264156415
> > 35%7CUnkn
> > >
> > own%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1
> > haWwi
> > >
> > LCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=tqzlcCmtSkUNxNzJ0Fr5o
> > CR3X2CTH
> > > 4dwUJS4NXt5KRc%3D&amp;reserved=0
> > > hu@xxxxxxx
> > > Signed-off-by: Richard Zhu <hongxing.zhu@xxxxxxx>
> > > Signed-off-by: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> > > ---
> > >  drivers/pci/controller/dwc/pci-imx6.c | 19 +++++++------------
> > >  1 file changed, 7 insertions(+), 12 deletions(-)
> > >
> > > diff --git a/drivers/pci/controller/dwc/pci-imx6.c
> > > b/drivers/pci/controller/dwc/pci-imx6.c
> > > index 6eddd0b5f628..537b8a2e0e3b 100644
> > > --- a/drivers/pci/controller/dwc/pci-imx6.c
> > > +++ b/drivers/pci/controller/dwc/pci-imx6.c
> > > @@ -670,8 +670,6 @@ static void imx6_pcie_clk_disable(struct imx6_pcie
> > > *imx6_pcie)
> > >
> > >  static void imx6_pcie_assert_core_reset(struct imx6_pcie *imx6_pcie)
> > > {
> > > -	struct device *dev = imx6_pcie->pci->dev;
> > > -
> > >  	switch (imx6_pcie->drvdata->variant) {
> > >  	case IMX7D:
> > >  	case IMX8MQ:
> > > @@ -702,14 +700,6 @@ static void imx6_pcie_assert_core_reset(struct
> > imx6_pcie *imx6_pcie)
> > >  		break;
> > >  	}
> > >
> > > -	if (imx6_pcie->vpcie && regulator_is_enabled(imx6_pcie->vpcie) > 0) {
> > > -		int ret = regulator_disable(imx6_pcie->vpcie);
> > > -
> > > -		if (ret)
> > > -			dev_err(dev, "failed to disable vpcie regulator: %d\n",
> > > -				ret);
> > > -	}
> > > -
> > >  	/* Some boards don't have PCIe reset GPIO. */
> > >  	if (gpio_is_valid(imx6_pcie->reset_gpio))
> > >  		gpio_set_value_cansleep(imx6_pcie->reset_gpio,
> > > @@ -722,7 +712,7 @@ static int imx6_pcie_deassert_core_reset(struct
> > imx6_pcie *imx6_pcie)
> > >  	struct device *dev = pci->dev;
> > >  	int ret, err;
> > >
> > > -	if (imx6_pcie->vpcie && !regulator_is_enabled(imx6_pcie->vpcie)) {
> > > +	if (imx6_pcie->vpcie) {
> > >  		ret = regulator_enable(imx6_pcie->vpcie);
> > >  		if (ret) {
> > >  			dev_err(dev, "failed to enable vpcie regulator: %d\n", @@
> > -796,7
> > > +786,7 @@ static int imx6_pcie_deassert_core_reset(struct imx6_pcie
> > *imx6_pcie)
> > >  	return 0;
> > >
> > >  err_clks:
> > > -	if (imx6_pcie->vpcie && regulator_is_enabled(imx6_pcie->vpcie) > 0) {
> > > +	if (imx6_pcie->vpcie) {
> > >  		ret = regulator_disable(imx6_pcie->vpcie);
> > >  		if (ret)
> > >  			dev_err(dev, "failed to disable vpcie regulator: %d\n", @@
> > -1023,6
> > > +1013,9 @@ static int imx6_pcie_suspend_noirq(struct device *dev)
> > >  		break;
> > >  	}
> > >
> > > +	if (imx6_pcie->vpcie)
> > > +		regulator_disable(imx6_pcie->vpcie);
> > 
> > This is a little bit ugly because imx6_pcie_suspend_noirq() and
> > imx6_pcie_resume_noirq() are not symmetric.
> > 
> > We call regulator_disable() directly here in imx6_pcie_suspend_noirq(), but the
> > corresponding regulator_enable() is buried in imx6_pcie_deassert_core_reset().
> > 
> > It would be nicer if the suspend and resume paths looked more similar, as
> > rockchip_pcie_suspend_noirq() and rockchip_pcie_resume_noirq() do, for
> > example.
> Yes, it is. The regulator_disable()/regulator_enable() are not invoked
>  symmetrically.
> 
> In the original codes, the regulator_disable() is contained in
> imx6_pcie_assert_core_reset() and with _eanbled() check.
> And regulator_enable() is embedded in imx6_pcie_deassert_core_reset().
> Both the regulator disable and enable are invoked in resume before.
> I move the regulator_disable to suspend mode refer to Lucas' comments.
> 
> Refer to the current situation, how about to move the regulator_disable()
>  out of imx6_pcie_deassert_core_reset() too?

I think it's important to tweak this closer to the pattern of other
drivers like exynos and rockchip.

exynos_pcie_resume_noirq() calls exynos_pcie_host_init().
imx6_pcie_resume_noirq() is VERY similar to imx6_pcie_host_init(), and
I think it should call imx6_pcie_host_init() instead of duplicating
that code.  The only real difference is that imx6_pcie_host_init()
calls imx6_setup_phy_mpll() while imx6_pcie_resume_noirq() does not.
I suspect that's a bug.

I don't think "regulator_enable(imx6_pcie->vpcie)" really has anything
to do with imx6_pcie_deassert_core_reset() and it should be moved out.
The typical pattern is to enable regulators in *_probe() and
*_resume_noirq() and disable them in *_remove() and *_suspend_noirq().
I think you should do the same unless there's some reason you can't.

Can you give this a try, starting with this v12 posting?

Bjorn