[PATCH] usb: core: hub: fix race condition about TRSMRCY of resume

[PATCH] usb: core: hub: fix race condition about TRSMRCY of resume

[Chunfeng Yun]

This may happen if the port becomes resume status exactly
when usb_port_resume() gets port status, it still need provide
a TRSMCRY time before access the device.

Reported-by: Tianping Fang <tianping.fang@mediatek.com>
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
---
 drivers/usb/core/hub.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index b2bc4b7c4289..fc7d6cdacf16 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -3642,9 +3642,6 @@ int usb_port_resume(struct usb_device *udev, pm_message_t msg)
 		 * sequence.
 		 */
 		status = hub_port_status(hub, port1, &portstatus, &portchange);
-
-		/* TRSMRCY = 10 msec */
-		msleep(10);
 	}
 
  SuspendCleared:
@@ -3659,6 +3656,9 @@ int usb_port_resume(struct usb_device *udev, pm_message_t msg)
 				usb_clear_port_feature(hub->hdev, port1,
 						USB_PORT_FEAT_C_SUSPEND);
 		}
+
+		/* TRSMRCY = 10 msec */
+		msleep(10);
 	}
 
 	if (udev->persist_enabled)
-- 
2.18.0

Re: [PATCH] usb: core: hub: fix race condition about TRSMRCY of resume

[Greg Kroah-Hartman]

On Tue, May 11, 2021 at 06:15:22PM +0800, Chunfeng Yun wrote:
> This may happen if the port becomes resume status exactly
> when usb_port_resume() gets port status, it still need provide
> a TRSMCRY time before access the device.
> 
> Reported-by: Tianping Fang <tianping.fang@mediatek.com>
> Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
> ---
>  drivers/usb/core/hub.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)

Does this need to go to any older/stable kernels?  If so, how far back?

thanks,

greg k-h

Re: [PATCH] usb: core: hub: fix race condition about TRSMRCY of resume

[Chunfeng Yun]

On Tue, 2021-05-11 at 13:27 +0200, Greg Kroah-Hartman wrote:
> On Tue, May 11, 2021 at 06:15:22PM +0800, Chunfeng Yun wrote:
> > This may happen if the port becomes resume status exactly
> > when usb_port_resume() gets port status, it still need provide
> > a TRSMCRY time before access the device.
> > 
> > Reported-by: Tianping Fang <tianping.fang@mediatek.com>
> > Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
> > ---
> >  drivers/usb/core/hub.c | 6 +++---
> >  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> Does this need to go to any older/stable kernels?  If so, how far back?

The flow is changed by 
"b01b03f3ad82 USB: add new routine for checking port-resume type", but
it will cause conflict with
"ad493e5e5805 usb: add usb port auto power off mechanism"
 
> 
> thanks,
> 
> greg k-h

Re: [PATCH] usb: core: hub: fix race condition about TRSMRCY of resume

[Alan Stern]

On Tue, May 11, 2021 at 06:15:22PM +0800, Chunfeng Yun wrote:
> This may happen if the port becomes resume status exactly
> when usb_port_resume() gets port status, it still need provide
> a TRSMCRY time before access the device.
> 
> Reported-by: Tianping Fang <tianping.fang@mediatek.com>
> Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>

This should also say:

CC: <stable@vger.kernel.org>

> ---
>  drivers/usb/core/hub.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
> index b2bc4b7c4289..fc7d6cdacf16 100644
> --- a/drivers/usb/core/hub.c
> +++ b/drivers/usb/core/hub.c
> @@ -3642,9 +3642,6 @@ int usb_port_resume(struct usb_device *udev, pm_message_t msg)
>  		 * sequence.
>  		 */
>  		status = hub_port_status(hub, port1, &portstatus, &portchange);
> -
> -		/* TRSMRCY = 10 msec */
> -		msleep(10);
>  	}
>  
>   SuspendCleared:
> @@ -3659,6 +3656,9 @@ int usb_port_resume(struct usb_device *udev, pm_message_t msg)
>  				usb_clear_port_feature(hub->hdev, port1,
>  						USB_PORT_FEAT_C_SUSPEND);
>  		}
> +
> +		/* TRSMRCY = 10 msec */
> +		msleep(10);
>  	}
>  
>  	if (udev->persist_enabled)

Acked-by: Alan Stern <stern@rowland.harvard.edu>

Re: [PATCH] usb: core: hub: fix race condition about TRSMRCY of resume

[Chunfeng Yun]

On Tue, 2021-05-11 at 12:39 -0400, Alan Stern wrote:
> On Tue, May 11, 2021 at 06:15:22PM +0800, Chunfeng Yun wrote:
> > This may happen if the port becomes resume status exactly
> > when usb_port_resume() gets port status, it still need provide
> > a TRSMCRY time before access the device.
> > 
> > Reported-by: Tianping Fang <tianping.fang@mediatek.com>
> > Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
> 
> This should also say:
> 
> CC: <stable@vger.kernel.org>

OK, thanks

> 
> > ---
> >  drivers/usb/core/hub.c | 6 +++---
> >  1 file changed, 3 insertions(+), 3 deletions(-)
> > 
> > diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
> > index b2bc4b7c4289..fc7d6cdacf16 100644
> > --- a/drivers/usb/core/hub.c
> > +++ b/drivers/usb/core/hub.c
> > @@ -3642,9 +3642,6 @@ int usb_port_resume(struct usb_device *udev, pm_message_t msg)
> >  		 * sequence.
> >  		 */
> >  		status = hub_port_status(hub, port1, &portstatus, &portchange);
> > -
> > -		/* TRSMRCY = 10 msec */
> > -		msleep(10);
> >  	}
> >  
> >   SuspendCleared:
> > @@ -3659,6 +3656,9 @@ int usb_port_resume(struct usb_device *udev, pm_message_t msg)
> >  				usb_clear_port_feature(hub->hdev, port1,
> >  						USB_PORT_FEAT_C_SUSPEND);
> >  		}
> > +
> > +		/* TRSMRCY = 10 msec */
> > +		msleep(10);
> >  	}
> >  
> >  	if (udev->persist_enabled)
> 
> Acked-by: Alan Stern <stern@rowland.harvard.edu>