| 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: <20220516150445.GB19209@hu-pkondeti-hyd.qualcomm.com>
Date: Mon, 16 May 2022 20:34:45 +0530
From: Pavan Kondeti <quic_pkondeti@...cinc.com>
To: Pavan Kondeti <quic_pkondeti@...cinc.com>
CC: Krishna Kurapati PSSNV <quic_kriskura@...cinc.com>,
Matthias Kaehlcke <mka@...omium.org>,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
Rob Herring <robh+dt@...nel.org>,
Andy Gross <agross@...nel.org>,
"Bjorn Andersson" <bjorn.andersson@...aro.org>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Felipe Balbi <balbi@...nel.org>,
Stephen Boyd <swboyd@...omium.org>,
Doug Anderson <dianders@...omium.org>,
Mathias Nyman <mathias.nyman@...el.com>,
<devicetree@...r.kernel.org>, <linux-arm-msm@...r.kernel.org>,
<linux-usb@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
<linux-pm@...r.kernel.org>, <quic_ppratap@...cinc.com>,
<quic_vpulyala@...cinc.com>,
Sandeep Maheswaram <quic_c_sanm@...cinc.com>
Subject: Re: [v16 2/5] usb: dwc3: core: Host wake up support from system
suspend
On Mon, May 16, 2022 at 10:13:27AM +0530, Pavan Kondeti wrote:
> On Fri, May 13, 2022 at 09:28:16AM +0530, Krishna Kurapati PSSNV wrote:
> >
> > On 5/13/2022 4:10 AM, Matthias Kaehlcke wrote:
> > >On Thu, May 12, 2022 at 11:53:19PM +0530, Krishna Kurapati wrote:
> > >>From: Sandeep Maheswaram <quic_c_sanm@...cinc.com>
> > >>
> > >>During suspend read the status of all port and set hs phy mode
> > >>based on current speed. Use this hs phy mode to configure wakeup
> > >>interrupts in qcom glue driver.
> > >>
> > >>Check wakeup-source property for dwc3 core node to set the
> > >>wakeup capability. Drop the device_init_wakeup call from
> > >>runtime suspend and resume.
> > >>
> > >>Also check during suspend if any wakeup capable devices are
> > >>connected to the controller (directly or through hubs), if there
> > >>are none set a flag to indicate that the PHY is powered
> > >>down during suspend.
> > >>
> > >>Signed-off-by: Sandeep Maheswaram <quic_c_sanm@...cinc.com>
> > >>Signed-off-by: Krishna Kurapati <quic_kriskura@...cinc.com>
> > >>---
> > >> drivers/usb/dwc3/core.c | 30 +++++++++++++++++-------------
> > >> drivers/usb/dwc3/core.h | 4 ++++
> > >> drivers/usb/dwc3/host.c | 24 ++++++++++++++++++++++++
> > >> 3 files changed, 45 insertions(+), 13 deletions(-)
> > >>
> > >>diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> > >>index 01115df..8bcabc5 100644
> > >>--- a/drivers/usb/dwc3/core.c
> > >>+++ b/drivers/usb/dwc3/core.c
> > >>@@ -1785,6 +1785,7 @@ static int dwc3_probe(struct platform_device *pdev)
> > >> platform_set_drvdata(pdev, dwc);
> > >> dwc3_cache_hwparams(dwc);
> > >>+ device_init_wakeup(&pdev->dev, of_property_read_bool(dev->of_node, "wakeup-source"));
> > >> spin_lock_init(&dwc->lock);
> > >> mutex_init(&dwc->mutex);
> > >>@@ -1946,10 +1947,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > >> dwc3_core_exit(dwc);
> > >> break;
> > >> case DWC3_GCTL_PRTCAP_HOST:
> > >>- if (!PMSG_IS_AUTO(msg)) {
> > >>- dwc3_core_exit(dwc);
> > >>- break;
> > >>- }
> > >>+ dwc3_check_phy_speed_mode(dwc);
> > >> /* Let controller to suspend HSPHY before PHY driver suspends */
> > >> if (dwc->dis_u2_susphy_quirk ||
> > >>@@ -1965,6 +1963,15 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg)
> > >> phy_pm_runtime_put_sync(dwc->usb2_generic_phy);
> > >> phy_pm_runtime_put_sync(dwc->usb3_generic_phy);
> > >>+
> > >>+ if (!PMSG_IS_AUTO(msg)) {
> > >>+ if (device_may_wakeup(dwc->dev))
> > >I think this should be device_can_wakeup(), i.e. hardware capability instead of
> > >device policy. A drawback of powering the PHYs off is that it causes a high
> > >power consumption of certain peripherals if VBUS is still supplied, so this
> > >should be limited to platforms where the PHYs must be powered off (using wakeup
> > >capability as a proxy for now).
> > Thnaks Mathias for the review. Will make this change in the next patchset.
> > >>+ dwc->phy_power_off = false;
> > >>+ else {
> > >>+ dwc->phy_power_off = true;
> > >>+ dwc3_core_exit(dwc);
> > >>+ }
> > >>+ }
> > >> break;
> > >> case DWC3_GCTL_PRTCAP_OTG:
> > >> /* do nothing during runtime_suspend */
> > >>@@ -2008,11 +2015,12 @@ static int dwc3_resume_common(struct dwc3 *dwc, pm_message_t msg)
> > >> break;
> > >> case DWC3_GCTL_PRTCAP_HOST:
> > >> if (!PMSG_IS_AUTO(msg)) {
> > >>- ret = dwc3_core_init_for_resume(dwc);
> > >>- if (ret)
> > >>- return ret;
> > >>- dwc3_set_prtcap(dwc, DWC3_GCTL_PRTCAP_HOST);
> > >>- break;
> > >>+ if (dwc->phy_power_off) {
> > >>+ ret = dwc3_core_init_for_resume(dwc);
> > >>+ if (ret)
> > >>+ return ret;
> > >>+ dwc3_set_prtcap(dwc, DWC3_GCTL_PRTCAP_HOST);
> > >>+ }
> > >> }
> > >> /* Restore GUSB2PHYCFG bits that were modified in suspend */
> > >> reg = dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0));
> > >>@@ -2084,8 +2092,6 @@ static int dwc3_runtime_suspend(struct device *dev)
> > >> if (ret)
> > >> return ret;
> > >>- device_init_wakeup(dev, true);
> > >>-
> > >> return 0;
> > >> }
> > >>@@ -2094,8 +2100,6 @@ static int dwc3_runtime_resume(struct device *dev)
> > >> struct dwc3 *dwc = dev_get_drvdata(dev);
> > >> int ret;
> > >>- device_init_wakeup(dev, false);
> > >>-
> > >> ret = dwc3_resume_common(dwc, PMSG_AUTO_RESUME);
> > >> if (ret)
> > >> return ret;
> > >>diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
> > >>index 81c486b..37397a8 100644
> > >>--- a/drivers/usb/dwc3/core.h
> > >>+++ b/drivers/usb/dwc3/core.h
> > >>@@ -1155,6 +1155,9 @@ struct dwc3 {
> > >> bool phys_ready;
> > >>+ unsigned int hs_phy_mode;
> > >>+ bool phy_power_off;
> > >>+
> > >> struct ulpi *ulpi;
> > >> bool ulpi_ready;
> > >>@@ -1539,6 +1542,7 @@ int dwc3_core_soft_reset(struct dwc3 *dwc);
> > >> #if IS_ENABLED(CONFIG_USB_DWC3_HOST) || IS_ENABLED(CONFIG_USB_DWC3_DUAL_ROLE)
> > >> int dwc3_host_init(struct dwc3 *dwc);
> > >> void dwc3_host_exit(struct dwc3 *dwc);
> > >>+void dwc3_check_phy_speed_mode(struct dwc3 *dwc);
> > >> #else
> > >> static inline int dwc3_host_init(struct dwc3 *dwc)
> > >> { return 0; }
> > >>diff --git a/drivers/usb/dwc3/host.c b/drivers/usb/dwc3/host.c
> > >>index f56c30c..e19b40a 100644
> > >>--- a/drivers/usb/dwc3/host.c
> > >>+++ b/drivers/usb/dwc3/host.c
> > >>@@ -12,6 +12,7 @@
> > >> #include <linux/platform_device.h>
> > >> #include "core.h"
> > >>+#include "../host/xhci.h"
> > >> static void dwc3_host_fill_xhci_irq_res(struct dwc3 *dwc,
> > >> int irq, char *name)
> > >>@@ -136,3 +137,26 @@ void dwc3_host_exit(struct dwc3 *dwc)
> > >> {
> > >> platform_device_unregister(dwc->xhci);
> > >> }
> > >>+
> > >>+void dwc3_check_phy_speed_mode(struct dwc3 *dwc)
> > >>+{
> > >>+ int i, num_ports;
> > >>+ u32 reg;
> > >>+ struct usb_hcd *hcd = platform_get_drvdata(dwc->xhci);
> > >>+ struct xhci_hcd *xhci_hcd = hcd_to_xhci(hcd);
> > >>+
> > >>+ dwc->hs_phy_mode = 0;
> > >>+
> > >>+ reg = readl(&xhci_hcd->cap_regs->hcs_params1);
> > >>+
> > >>+ num_ports = HCS_MAX_PORTS(reg);
> > >>+ for (i = 0; i < num_ports; i++) {
> > >>+ reg = readl(&xhci_hcd->op_regs->port_status_base + i * NUM_PORT_REGS);
> > >>+ if (reg & PORT_PE) {
> > >>+ if (DEV_HIGHSPEED(reg) || DEV_FULLSPEED(reg))
> > >>+ dwc->hs_phy_mode |= PHY_MODE_USB_HOST_HS;
> > >>+ else if (DEV_LOWSPEED(reg))
> > >>+ dwc->hs_phy_mode |= PHY_MODE_USB_HOST_LS;
> > >>+ }
> > >>+ }
> > >>+}
> > >I anticipate that it might raise concerns from maintainers that
> > >dwc3_check_phy_speed_mode() accesses xHCI data structures and
> > >registers directly. Could there be a generic HCD API that provides
> > >this functionality (if implemented by the specific HCD)?
> >
> > Hi Mathias, we are not sure if there is any such API present currently.
> >
> > Hi Alan, can you help suggest any API (if present) that we can reuse here to
> > avoid
> >
> > xhci registers and structs here in dwc3.
> >
>
> We can probably do something like below to query the speed. This avoids adding
> another API and does not touch the underlying registers.
>
> Pls define enum usb_device_speed usb2_speed in dwc3 structure.
>
> diff --git a/drivers/usb/dwc3/host.c b/drivers/usb/dwc3/host.c
> index f29a264..fed1c58 100644
> --- a/drivers/usb/dwc3/host.c
> +++ b/drivers/usb/dwc3/host.c
> @@ -9,9 +9,29 @@
>
> #include <linux/acpi.h>
> #include <linux/platform_device.h>
> +#include <linux/usb.h>
>
> #include "core.h"
>
> +void dwc3_update_hs_phy_speed(struct dwc3 *dwc)
> +{
> + struct usb_hcd *hcd = platform_get_drvdata(dwc->xhci);
> + struct usb_device *udev;
> +
> + /*
> + * It is possible to query the speed of all children of
> + * USB2.0 root hub via usb_hub_for_each_child(). DWC3 code
> + * currently supports only 1 port per controller. So
> + * this is sufficient.
> + */
> + udev = usb_hub_find_child(hcd->self.root_hub, 1);
> +
> + if (udev)
> + dwc->usb2_speed = udev->speed;
> + else
> + dwc->usb2_speed = USB_SPEED_UNKNOWN;
> +}
> +
> static int dwc3_host_get_irq(struct dwc3 *dwc)
> {
> struct platform_device *dwc3_pdev = to_platform_device(dwc->dev);
>
>
I am also thinking why dwc core needs to cache usb2_speed since dwc3-qcom glue
driver is the only sole user. We also require it only during suspend and does
not bother about dwc::usb2_speed correctness outside suspend. Lets move this
to dwc3-qcom suspend routines where we have to rely on USB2 speed for
configuring the D+/D- interrupt.
Thanks,
Pavan
Powered by blists - more mailing lists