| 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
| ||
|
Date: Tue, 7 Jul 2015 09:12:17 +0000
From: Phil Edworthy <phil.edworthy@...esas.com>
To: Laurent Pinchart <laurent.pinchart@...asonboard.com>
CC: Yoshihiro Shimoda <yoshihiro.shimoda.uh@...esas.com>,
Kuninori Morimoto <kuninori.morimoto.gx@...esas.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Felipe Balbi <balbi@...com>,
Ulrich Hecht <ulrich.hecht@...il.com>,
Kishon Vijay Abraham I <kishon@...com>,
"Sergei Shtylyov" <sergei.shtylyov@...entembedded.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-usb@...r.kernel.org" <linux-usb@...r.kernel.org>,
"linux-sh@...r.kernel.org" <linux-sh@...r.kernel.org>
Subject: RE: [PATCH v3] usb: renesas_usbhs: Allow an OTG PHY driver to provide
VBUS
Hi Laurent,
On 06 July 2015 09:20, Laurent wrote:
> Hi Phil,
>
> Thank you for the patch.
Thanks for your review!
> On Thursday 02 July 2015 11:26:33 Phil Edworthy wrote:
> > These changes allow a PHY driver to trigger a VBUS interrupt and
> > to provide the value of VBUS.
> >
> > Signed-off-by: Phil Edworthy <phil.edworthy@...esas.com>
> >
> > ---
> > v3:
> > - Changed how indirect vbus is plumbed in.
> > - Removed unnecessary (void) on call to otg_set_peripheral.
> > - Moved code that connects to bus through transceiver so it
> > is before setting gpriv->driver.
> >
> > v2:
> > - vbus variables changed from int to bool.
> > - dev_info() changed to dev_err()
> > ---
> > drivers/usb/renesas_usbhs/common.c | 8 +++++--
> > drivers/usb/renesas_usbhs/common.h | 2 ++
> > drivers/usb/renesas_usbhs/mod_gadget.c | 38
> +++++++++++++++++++++++++++++++
> > 3 files changed, 46 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/usb/renesas_usbhs/common.c
> > b/drivers/usb/renesas_usbhs/common.c index 0f7e850..15e67d8 100644
> > --- a/drivers/usb/renesas_usbhs/common.c
> > +++ b/drivers/usb/renesas_usbhs/common.c
> > @@ -379,7 +379,10 @@ static void usbhsc_hotplug(struct usbhs_priv *priv)
> > /*
> > * get vbus status from platform
> > */
> > - enable = usbhs_platform_call(priv, get_vbus, pdev);
> > + if (priv->vbus_is_indirect)
> > + enable = priv->vbus_indirect_value;
> > + else
> > + enable = usbhs_platform_call(priv, get_vbus, pdev);
> >
> > /*
> > * get id from platform
> > @@ -662,7 +665,8 @@ static int usbhs_probe(struct platform_device *pdev)
> > pm_runtime_enable(&pdev->dev);
> > if (!usbhsc_flags_has(priv, USBHSF_RUNTIME_PWCTRL)) {
> > usbhsc_power_ctrl(priv, 1);
> > - usbhs_mod_autonomy_mode(priv);
> > + if (!priv->vbus_is_indirect)
>
> Isn't this racy ? The vbus_session operation is called by transceivers from
> interrupt handlers or work queues. Do we have a guarantee that it will be
> called at least once before this code is reached ? Please see below for a
> possible way to fix this.
You are right, it is racy... :(
> > + usbhs_mod_autonomy_mode(priv);
> > }
> >
> > /*
> > diff --git a/drivers/usb/renesas_usbhs/common.h
> > b/drivers/usb/renesas_usbhs/common.h index 8c5fc12..478700c 100644
> > --- a/drivers/usb/renesas_usbhs/common.h
> > +++ b/drivers/usb/renesas_usbhs/common.h
> > @@ -255,6 +255,8 @@ struct usbhs_priv {
> > struct renesas_usbhs_driver_param dparam;
> >
> > struct delayed_work notify_hotplug_work;
> > + bool vbus_is_indirect;
> > + bool vbus_indirect_value;
> > struct platform_device *pdev;
> >
> > struct extcon_dev *edev;
> > diff --git a/drivers/usb/renesas_usbhs/mod_gadget.c
> > b/drivers/usb/renesas_usbhs/mod_gadget.c index dc2aa32..93ad4ea 100644
> > --- a/drivers/usb/renesas_usbhs/mod_gadget.c
> > +++ b/drivers/usb/renesas_usbhs/mod_gadget.c
> > @@ -21,6 +21,7 @@
> > #include <linux/platform_device.h>
> > #include <linux/usb/ch9.h>
> > #include <linux/usb/gadget.h>
> > +#include <linux/usb/otg.h>
> > #include "common.h"
> >
> > /*
> > @@ -50,6 +51,7 @@ struct usbhsg_gpriv {
> > int uep_size;
> >
> > struct usb_gadget_driver *driver;
> > + struct usb_phy *transceiver;
> >
> > u32 status;
> > #define USBHSG_STATUS_STARTED (1 << 0)
> > @@ -882,12 +884,25 @@ static int usbhsg_gadget_start(struct usb_gadget
> > *gadget, {
> > struct usbhsg_gpriv *gpriv = usbhsg_gadget_to_gpriv(gadget);
> > struct usbhs_priv *priv = usbhsg_gpriv_to_priv(gpriv);
> > + struct device *dev = usbhs_priv_to_dev(priv);
> > + int ret;
> >
> > if (!driver ||
> > !driver->setup ||
> > driver->max_speed < USB_SPEED_FULL)
> > return -EINVAL;
> >
> > + /* connect to bus through transceiver */
> > + if (!IS_ERR_OR_NULL(gpriv->transceiver)) {
> > + ret = otg_set_peripheral(gpriv->transceiver->otg,
> > + &gpriv->gadget);
> > + if (ret) {
> > + dev_err(dev, "%s: can't bind to transceiver\n",
> > + gpriv->gadget.name);
> > + return ret;
> > + }
> > + }
> > +
>
> How is this change related to the topic ? I'm not too familiar with this
> driver so I might just have missed something, but then others could as well,
> and the commit message could do with a bit more information.
I'm not sure what you mean, this code is needed to tell the phy driver
if it is in host or gadget mode. I would have thought this is clear as it calls
otg_set_peripheral.
> > /* first hook up the driver ... */
> > gpriv->driver = driver;
> >
> > @@ -900,6 +915,10 @@ static int usbhsg_gadget_stop(struct usb_gadget
> > *gadget) struct usbhs_priv *priv = usbhsg_gpriv_to_priv(gpriv);
> >
> > usbhsg_try_stop(priv, USBHSG_STATUS_REGISTERD);
> > +
> > + if (!IS_ERR_OR_NULL(gpriv->transceiver))
> > + otg_set_peripheral(gpriv->transceiver->otg, NULL);
> > +
> > gpriv->driver = NULL;
> >
> > return 0;
> > @@ -947,12 +966,27 @@ static int usbhsg_set_selfpowered(struct usb_gadget
> > *gadget, int is_self) return 0;
> > }
> >
> > +static int usbhsg_vbus_session(struct usb_gadget *gadget, int is_active)
> > +{
> > + struct usbhsg_gpriv *gpriv = usbhsg_gadget_to_gpriv(gadget);
> > + struct usbhs_priv *priv = usbhsg_gpriv_to_priv(gpriv);
> > + struct platform_device *pdev = usbhs_priv_to_pdev(priv);
> > +
> > + priv->vbus_is_indirect = 1;
>
> vbus_is_indirect is never reset to 0. Shouldn't you set it to 1 (or rather to
> true) in usbhs_mod_gadget_probe() if an external transceiver is found ?
Yes, I see what you say.
> What would you think of implementing and wiring up the get_vbus operation in
> usbhs_mod_gadget_probe() in the same way that
> usbhs_mod_autonomy_mode() does,
> and return the value of vbus_indirect_value there ? You could then get rid of
> the vbus_is_indirect field and move the vbus_indirect_value field to struct
> usbhsg_gpriv, removing the need to modify
> drivers/usb/renesas_usbhs/common.c.
> It should also solve the race condition mentioned above.
Ok, I'll have a look at that.
> > + priv->vbus_indirect_value = !!is_active;
> > +
> > + renesas_usbhs_call_notify_hotplug(pdev);
> > +
> > + return 0;
> > +}
> > +
> > static const struct usb_gadget_ops usbhsg_gadget_ops = {
> > .get_frame = usbhsg_get_frame,
> > .set_selfpowered = usbhsg_set_selfpowered,
> > .udc_start = usbhsg_gadget_start,
> > .udc_stop = usbhsg_gadget_stop,
> > .pullup = usbhsg_pullup,
> > + .vbus_session = usbhsg_vbus_session,
> > };
> >
> > static int usbhsg_start(struct usbhs_priv *priv)
> > @@ -994,6 +1028,10 @@ int usbhs_mod_gadget_probe(struct usbhs_priv
> *priv)
> > goto usbhs_mod_gadget_probe_err_gpriv;
> > }
> >
> > + gpriv->transceiver = usb_get_phy(USB_PHY_TYPE_UNDEFINED);
> > + dev_info(dev, "%s transceiver found\n",
> > + gpriv->transceiver ? "" : "No");
> > +
> > /*
> > * CAUTION
> > *
>
> --
> Regards,
>
> Laurent Pinchart
Thanks
Phil
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists