| 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
| ||
|
Message-ID: <20240827165826.moe6cnemeheos6jn@thinkpad>
Date: Tue, 27 Aug 2024 22:28:26 +0530
From: Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
To: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
Cc: Qiang Yu <quic_qianyu@...cinc.com>, vkoul@...nel.org, kishon@...nel.org,
robh@...nel.org, andersson@...nel.org, konradybcio@...nel.org,
krzk+dt@...nel.org, conor+dt@...nel.org, mturquette@...libre.com,
sboyd@...nel.org, abel.vesa@...aro.org, quic_msarkar@...cinc.com,
quic_devipriy@...cinc.com, kw@...ux.com, lpieralisi@...nel.org,
neil.armstrong@...aro.org, linux-arm-msm@...r.kernel.org,
linux-phy@...ts.infradead.org, linux-kernel@...r.kernel.org,
linux-pci@...r.kernel.org, devicetree@...r.kernel.org,
linux-clk@...r.kernel.org
Subject: Re: [PATCH 8/8] PCI: qcom: Add support to PCIe slot power supplies
On Tue, Aug 27, 2024 at 02:44:09PM +0300, Dmitry Baryshkov wrote:
> On Tue, 27 Aug 2024 at 09:36, Qiang Yu <quic_qianyu@...cinc.com> wrote:
> >
> > On platform x1e80100 QCP, PCIe3 is a standard x8 form factor. Hence, add
> > support to use 3.3v, 3.3v aux and 12v regulators.
>
> First of all, I don't see corresponding bindings change.
>
> Second, these supplies power up the slot, not the host controller
> itself. As such these supplies do not belong to the host controller
> entry. Please consider using the pwrseq framework instead.
>
Indeed. For legacy reasons, slot power supplies were populated in the host
bridge node itself until recently Rob started objecting it [1]. And it makes
real sense to put these supplies in the root port node and handle them in the
relevant driver.
I'm still evaluating whether the handling should be done in the portdrv or
pwrctl driver, but haven't reached the conclusion. Pwrctl seems to be the ideal
choice, but I see a few issues related to handling the OF node for the root
port.
Hope I'll come to a conclusion in the next few days and will update this thread.
- Mani
[1] https://lore.kernel.org/lkml/20240604235806.GA1903493-robh@kernel.org/
> >
> > Signed-off-by: Qiang Yu <quic_qianyu@...cinc.com>
> > ---
> > drivers/pci/controller/dwc/pcie-qcom.c | 52 +++++++++++++++++++++++++-
> > 1 file changed, 50 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/pci/controller/dwc/pcie-qcom.c b/drivers/pci/controller/dwc/pcie-qcom.c
> > index 6f953e32d990..59fb415dfeeb 100644
> > --- a/drivers/pci/controller/dwc/pcie-qcom.c
> > +++ b/drivers/pci/controller/dwc/pcie-qcom.c
> > @@ -248,6 +248,8 @@ struct qcom_pcie_cfg {
> > bool no_l0s;
> > };
> >
> > +#define QCOM_PCIE_SLOT_MAX_SUPPLIES 3
> > +
> > struct qcom_pcie {
> > struct dw_pcie *pci;
> > void __iomem *parf; /* DT parf */
> > @@ -260,6 +262,7 @@ struct qcom_pcie {
> > struct icc_path *icc_cpu;
> > const struct qcom_pcie_cfg *cfg;
> > struct dentry *debugfs;
> > + struct regulator_bulk_data slot_supplies[QCOM_PCIE_SLOT_MAX_SUPPLIES];
> > bool suspended;
> > bool use_pm_opp;
> > };
> > @@ -1174,6 +1177,41 @@ static int qcom_pcie_link_up(struct dw_pcie *pci)
> > return !!(val & PCI_EXP_LNKSTA_DLLLA);
> > }
> >
> > +static int qcom_pcie_enable_slot_supplies(struct qcom_pcie *pcie)
> > +{
> > + struct dw_pcie *pci = pcie->pci;
> > + int ret;
> > +
> > + ret = regulator_bulk_enable(ARRAY_SIZE(pcie->slot_supplies),
> > + pcie->slot_supplies);
> > + if (ret < 0)
> > + dev_err(pci->dev, "Failed to enable slot regulators\n");
> > +
> > + return ret;
> > +}
> > +
> > +static void qcom_pcie_disable_slot_supplies(struct qcom_pcie *pcie)
> > +{
> > + regulator_bulk_disable(ARRAY_SIZE(pcie->slot_supplies),
> > + pcie->slot_supplies);
> > +}
> > +
> > +static int qcom_pcie_get_slot_supplies(struct qcom_pcie *pcie)
> > +{
> > + struct dw_pcie *pci = pcie->pci;
> > + int ret;
> > +
> > + pcie->slot_supplies[0].supply = "vpcie12v";
> > + pcie->slot_supplies[1].supply = "vpcie3v3";
> > + pcie->slot_supplies[2].supply = "vpcie3v3aux";
> > + ret = devm_regulator_bulk_get(pci->dev, ARRAY_SIZE(pcie->slot_supplies),
> > + pcie->slot_supplies);
> > + if (ret < 0)
> > + dev_err(pci->dev, "Failed to get slot regulators\n");
> > +
> > + return ret;
> > +}
> > +
> > static int qcom_pcie_host_init(struct dw_pcie_rp *pp)
> > {
> > struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
> > @@ -1182,10 +1220,14 @@ static int qcom_pcie_host_init(struct dw_pcie_rp *pp)
> >
> > qcom_ep_reset_assert(pcie);
> >
> > - ret = pcie->cfg->ops->init(pcie);
> > + ret = qcom_pcie_enable_slot_supplies(pcie);
> > if (ret)
> > return ret;
> >
> > + ret = pcie->cfg->ops->init(pcie);
> > + if (ret)
> > + goto err_disable_slot;
> > +
> > ret = phy_set_mode_ext(pcie->phy, PHY_MODE_PCIE, PHY_MODE_PCIE_RC);
> > if (ret)
> > goto err_deinit;
> > @@ -1216,7 +1258,8 @@ static int qcom_pcie_host_init(struct dw_pcie_rp *pp)
> > phy_power_off(pcie->phy);
> > err_deinit:
> > pcie->cfg->ops->deinit(pcie);
> > -
> > +err_disable_slot:
> > + qcom_pcie_disable_slot_supplies(pcie);
> > return ret;
> > }
> >
> > @@ -1228,6 +1271,7 @@ static void qcom_pcie_host_deinit(struct dw_pcie_rp *pp)
> > qcom_ep_reset_assert(pcie);
> > phy_power_off(pcie->phy);
> > pcie->cfg->ops->deinit(pcie);
> > + qcom_pcie_disable_slot_supplies(pcie);
> > }
> >
> > static void qcom_pcie_host_post_init(struct dw_pcie_rp *pp)
> > @@ -1602,6 +1646,10 @@ static int qcom_pcie_probe(struct platform_device *pdev)
> > goto err_pm_runtime_put;
> > }
> >
> > + ret = qcom_pcie_get_slot_supplies(pcie);
> > + if (ret)
> > + goto err_pm_runtime_put;
> > +
> > ret = pcie->cfg->ops->get_resources(pcie);
> > if (ret)
> > goto err_pm_runtime_put;
> > --
> > 2.34.1
> >
>
>
> --
> With best wishes
> Dmitry
--
மணிவண்ணன் சதாசிவம்
Powered by blists - more mailing lists