Re: [PATCH v14 10/17] PCI: imx6: Turn off regulator when system is in suspend mode

[Lucas Stach <l.stach@pengutronix.de>]

Am Mittwoch, dem 13.07.2022 um 10:56 +0000 schrieb Hongxing Zhu:
> > -----Original Message-----
> > From: Lucas Stach <l.stach@pengutronix.de>
> > Sent: 2022年7月13日 16:34
> > To: Hongxing Zhu <hongxing.zhu@nxp.com>; bhelgaas@google.com;
> > robh+dt@kernel.org; broonie@kernel.org; lorenzo.pieralisi@arm.com;
> > festevam@gmail.com; francesco.dolcini@toradex.com
> > Cc: linux-pci@vger.kernel.org;
> > linux-arm-kernel@lists.infradead.org;
> > linux-kernel@vger.kernel.org; kernel@pengutronix.de; dl-linux-imx
> > <linux-imx@nxp.com>
> > Subject: Re: [PATCH v14 10/17] PCI: imx6: Turn off regulator when
> > system is in
> > suspend mode
> > 
> > Am Freitag, dem 01.07.2022 um 11:25 +0800 schrieb Richard Zhu:
> > > 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%7C361c9bf365b64895482808da6
> > 4aa7cca%
> > > 
> > 7C686ea1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C6379329806385373
> > 39%7CUnkn
> > > 
> > own%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1
> > haWwi
> > > 
> > LCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=A%2Fw6AFCddC0T%2FA
> > ocBT%2F7OJZ
> > > ddAeYPTbgeAxpHpPubkw%3D&amp;reserved=0
> > > hu@nxp.com
> > > Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
> > > Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
> > > ---
> > >  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 2b42c37f1617..f72eb609769b 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;
> > > 
> > > -	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",
> > 
> > The regulator really has nothing to do with the core reset. Please
> > move this
> > regulator enable into imx6_pcie_host_init().
> Hi Lucas:
> Thanks for your comments.
> Got that. Had done it in the 11/17 commit.
> 
> > 
> > > @@ -795,7 +785,7 @@ 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", @@
> > -1022,6
> > > +1012,9 @@ static int imx6_pcie_suspend_noirq(struct device *dev)
> > >  		break;
> > >  	}
> > > 
> > > +	if (imx6_pcie->vpcie)
> > > +		regulator_disable(imx6_pcie->vpcie);
> > > +
> > >  	return 0;
> > >  }
> > > 
> > > @@ -1268,6 +1261,8 @@ static void imx6_pcie_shutdown(struct
> > > platform_device *pdev)
> > > 
> > >  	/* bring down link, so bootloader gets clean state in
> > > case of reboot */
> > >  	imx6_pcie_assert_core_reset(imx6_pcie);
> > > +	if (imx6_pcie->vpcie)
> > > +		regulator_disable(imx6_pcie->vpcie);
> > 
> > This looks like a separate change, not mentioned in the commit
> > message.
> > I'm not sure if we should do this. Shutdown is supposed to just
> > stop the device,
> > which is already achieved by imx6_pcie_assert_core_reset().
> > 
> > If we would want to do a full cleanup here we would also need to
> > disable
> > clocks and get the reset GPIO into asserted state. I don't think we
> > want to do
> > all of this here.
> The regulator_disable() was contained in
> imx6_pcie_assert_core_reset() before.
> When refine the regulator usage, the regulator_disable() is moved out
> of
>  imx6_pcie_assert_core_reset().
> Based on the discussion [1] with Francesco.
> To keep the same behavior, the regulator_disable() is placed behind
> of
>  imx6_pcie_assert_core_reset() in imx6_pcie_shutdown() here.
> [1]
> https://patchwork.ozlabs.org/project/linux-pci/patch/1644290735-3797-6-git-send-email-hongxing.zhu@nxp.com/

I disagree with Francesco on this. This kind of half-done cleanup
doesn't help in any way. Either we care about doing a full cleanup
here, which would involve clocks and reset GPIOs, or we just care about
getting the core in a state where it survives a reboot, which frankly
was the original intention when I added the shutdown callback.

As the devices on the bus will see a full PERST reset via the reset
GPIO anyways, there is no harm in keeping them powered across the
reboot.

Regards,
Lucas


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel