| 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: <CAH=2NtzVEatke2LFyTFJDBgrQ07yz-AQgV7ubq+F6oxPmbKiZw@mail.gmail.com>
Date: Thu, 4 May 2023 22:40:14 +0530
From: Bhupesh Sharma <bhupesh.sharma@...aro.org>
To: Konrad Dybcio <konrad.dybcio@...aro.org>
Cc: linux-arm-msm@...r.kernel.org, devicetree@...r.kernel.org,
agross@...nel.org, andersson@...nel.org,
linux-kernel@...r.kernel.org, bhupesh.linux@...il.com,
robh+dt@...nel.org, linux-usb@...r.kernel.org,
krzysztof.kozlowski+dt@...aro.org, krzysztof.kozlowski@...aro.org
Subject: Re: [PATCH v3 3/5] usb: misc: eud: Add driver support for SM6115 / SM4250
On Thu, 4 May 2023 at 15:22, Konrad Dybcio <konrad.dybcio@...aro.org> wrote:
>
> On 4.05.2023 10:26, Bhupesh Sharma wrote:
> > Add SM6115 / SM4250 SoC EUD support in qcom_eud driver.
> >
> > On some SoCs (like the SM6115 / SM4250 SoC), the mode manager
> > needs to be accessed only via the secure world (through 'scm'
> > calls).
> >
> > Also, the enable bit inside 'tcsr_check_reg' needs to be set
> > first to set the eud in 'enable' mode on these SoCs.
> >
> > Since this difference comes from how the firmware is configured, so
> > the driver now relies on the presence of an extra boolean DT property
> > to identify if secure access is needed.
> >
> > Signed-off-by: Bhupesh Sharma <bhupesh.sharma@...aro.org>
> > ---
> > drivers/usb/misc/Kconfig | 1 +
> > drivers/usb/misc/qcom_eud.c | 66 ++++++++++++++++++++++++++++++++++---
> > 2 files changed, 62 insertions(+), 5 deletions(-)
> >
> > diff --git a/drivers/usb/misc/Kconfig b/drivers/usb/misc/Kconfig
> > index 99b15b77dfd5..fe1b5fec1dfc 100644
> > --- a/drivers/usb/misc/Kconfig
> > +++ b/drivers/usb/misc/Kconfig
> > @@ -147,6 +147,7 @@ config USB_APPLEDISPLAY
> > config USB_QCOM_EUD
> > tristate "QCOM Embedded USB Debugger(EUD) Driver"
> > depends on ARCH_QCOM || COMPILE_TEST
> > + select QCOM_SCM
> > select USB_ROLE_SWITCH
> > help
> > This module enables support for Qualcomm Technologies, Inc.
> > diff --git a/drivers/usb/misc/qcom_eud.c b/drivers/usb/misc/qcom_eud.c
> > index b7f13df00764..b4736edcc64c 100644
> > --- a/drivers/usb/misc/qcom_eud.c
> > +++ b/drivers/usb/misc/qcom_eud.c
> > @@ -5,12 +5,14 @@
> >
> > #include <linux/bitops.h>
> > #include <linux/err.h>
> > +#include <linux/firmware/qcom/qcom_scm.h>
> > #include <linux/interrupt.h>
> > #include <linux/io.h>
> > #include <linux/iopoll.h>
> > #include <linux/kernel.h>
> > #include <linux/module.h>
> > #include <linux/of.h>
> > +#include <linux/of_device.h>
> > #include <linux/platform_device.h>
> > #include <linux/slab.h>
> > #include <linux/sysfs.h>
> > @@ -30,15 +32,22 @@
> > #define EUD_INT_SAFE_MODE BIT(4)
> > #define EUD_INT_ALL (EUD_INT_VBUS | EUD_INT_SAFE_MODE)
> >
> > +struct eud_soc_cfg {
> > + u32 tcsr_check_offset;
> > +};
> Not sure if turning this into a struct is necessary.. can't
> we just store the offset, or do we expect more changes?
I can see future versions already supporting newer features, so I kept
it a struct for now.
> > +
> > struct eud_chip {
> > struct device *dev;
> > struct usb_role_switch *role_sw;
> > + const struct eud_soc_cfg *eud_cfg;
> > void __iomem *base;
> > void __iomem *mode_mgr;
> > unsigned int int_status;
> > int irq;
> > bool enabled;
> > bool usb_attached;
> > + bool secure_mode_enable;
> > + phys_addr_t secure_mode_mgr;
> > };
> >
> > static int enable_eud(struct eud_chip *priv)
> > @@ -46,7 +55,11 @@ static int enable_eud(struct eud_chip *priv)
> > writel(EUD_ENABLE, priv->base + EUD_REG_CSR_EUD_EN);
> > writel(EUD_INT_VBUS | EUD_INT_SAFE_MODE,
> > priv->base + EUD_REG_INT1_EN_MASK);
> > - writel(1, priv->mode_mgr + EUD_REG_EUD_EN2);
> > +
> > + if (priv->secure_mode_mgr)
> > + qcom_scm_io_writel(priv->secure_mode_mgr + EUD_REG_EUD_EN2, BIT(0));
> #define [field name] BIT(0)
Ok.
> > + else
> > + writel(1, priv->mode_mgr + EUD_REG_EUD_EN2);
> s/1/[field name]/
Ok.
> > return usb_role_switch_set_role(priv->role_sw, USB_ROLE_DEVICE);
> > }
> > @@ -54,7 +67,11 @@ static int enable_eud(struct eud_chip *priv)
> > static void disable_eud(struct eud_chip *priv)
> > {
> > writel(0, priv->base + EUD_REG_CSR_EUD_EN);
> > - writel(0, priv->mode_mgr + EUD_REG_EUD_EN2);
> > +
> > + if (priv->secure_mode_mgr)
> > + qcom_scm_io_writel(priv->secure_mode_mgr + EUD_REG_EUD_EN2, 0);
> > + else
> > + writel(0, priv->mode_mgr + EUD_REG_EUD_EN2);
> > }
> >
> > static ssize_t enable_show(struct device *dev,
> > @@ -178,12 +195,15 @@ static void eud_role_switch_release(void *data)
> > static int eud_probe(struct platform_device *pdev)
> > {
> > struct eud_chip *chip;
> > + struct resource *res;
> > + phys_addr_t tcsr_base, tcsr_check;
> > int ret;
> >
> > chip = devm_kzalloc(&pdev->dev, sizeof(*chip), GFP_KERNEL);
> > if (!chip)
> > return -ENOMEM;
> >
> > +
> ?
Oops, I will fix it in v4.
> > chip->dev = &pdev->dev;
> >
> > chip->role_sw = usb_role_switch_get(&pdev->dev);
> > @@ -200,9 +220,40 @@ static int eud_probe(struct platform_device *pdev)
> > if (IS_ERR(chip->base))
> > return PTR_ERR(chip->base);
> >
> > - chip->mode_mgr = devm_platform_ioremap_resource(pdev, 1);
> > - if (IS_ERR(chip->mode_mgr))
> > - return PTR_ERR(chip->mode_mgr);
> > + chip->secure_mode_enable = of_property_read_bool(chip->dev->of_node,
> > + "qcom,secure-mode-enable");
> If we map this region iff it's supposed to be used, we may just check
> for its presence and skip the additional property. Then, the address
> being non-NULL would invalidate the boolean property.
Bjorn requested during the review of the last version that we should not ioremap
the secure mode_mgr region. So, I followed this approach instead.
> > + /*
> > + * EUD block on a few Qualcomm SoCs need secure register access.
> > + * Check for the same.
> > + */
> > + if (chip->secure_mode_enable) {
> > + res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
> > + if (!res)
> > + return dev_err_probe(chip->dev, -ENODEV,
> > + "failed to get secure_mode_mgr reg base\n");
> > +
> > + chip->secure_mode_mgr = res->start;
> > + } else {
> > + chip->mode_mgr = devm_platform_ioremap_resource(pdev, 1);
> > + if (IS_ERR(chip->mode_mgr))
> > + return PTR_ERR(chip->mode_mgr);
> > + }
> > +
> > + /* Check for any SoC specific config data */
> > + chip->eud_cfg = of_device_get_match_data(&pdev->dev);
> > + if (chip->eud_cfg) {
> > + res = platform_get_resource(pdev, IORESOURCE_MEM, 2);
> I'd vouch to use _byname, in case we get some EUD impl that needs a
> different sort of a register set..
Sure, it makes sense.
> > + if (!res)
> > + return dev_err_probe(chip->dev, -ENODEV,
> > + "failed to get tcsr reg base\n");
> > +
> > + tcsr_base = res->start;
> > + tcsr_check = tcsr_base + chip->eud_cfg->tcsr_check_offset;
> > +
> > + ret = qcom_scm_io_writel(tcsr_check, BIT(0));
> s/BIT(0)/..
Ok.
Thanks,
Bhupesh
Powered by blists - more mailing lists