[PATCH] fix some hubs and hid devices at startup

[PATCH] fix some hubs and hid devices at startup

[Greg KH]

Hi,

Since the latest USB patch went in that added a proper delay to the
hub connection sequence, a lot of people have been reporting odd
problems with some USB hubs, keyboards, and mice.  I've seen this
problem too.  It is usually fixed by just replugging the device in, but
not all of the time.

The reason seems to be that usbmodules is getting called from the
hotplug package at the same time the hub driver is talking to the
device.  Since we don't have control endpoint locking in the kernel USB
code yet, this messes things up :)

The patch below, from Olaf Hering, seems to fix the problem.  Also just
removing the usbmodules binary seems to fix it for other people, but
that's not really a good fix.

Let me know what you think about this.

thanks,

greg k-h


Index: etc/hotplug/usb.agent
=================================RCS file: /cvsroot/linux-hotplug/admin/etc/hotplug/usb.agent,v
retrieving revision 1.19
diff -a -u -r1.19 usb.agent
--- etc/hotplug/usb.agent	18 Jan 2002 03:10:42 -0000	1.19
+++ etc/hotplug/usb.agent	11 Mar 2002 21:52:10 -0000
@@ -297,6 +297,9 @@
 case $ACTION in
 
 add)
+    # Let the usb subsystem finish talking to the device, before we do
+    sleep 3
+
     usb_convert_vars
 
     FOUNDúlse

_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: [PATCH] fix some hubs and hid devices at startup

[Greg KH]

On Mon, Mar 11, 2002 at 11:48:15PM +0100, Oliver Neukum wrote:
> 
> > The patch below, from Olaf Hering, seems to fix the problem.  Also just
> > removing the usbmodules binary seems to fix it for other people, but
> > that's not really a good fix.
> >
> > Let me know what you think about this.
> 
> It is voodoo, but it works.
> I trust it won't keep you from fixing the locking.

No, it will not :)
Thanks for reminding me to add this to the TODO list.

greg k-h

_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: [PATCH] fix some hubs and hid devices at startup

[Olaf Hering]

On Mon, Mar 11, Greg KH wrote:

> Hi,
> 
> Since the latest USB patch went in that added a proper delay to the
> hub connection sequence, a lot of people have been reporting odd
> problems with some USB hubs, keyboards, and mice.  I've seen this
> problem too.  It is usually fixed by just replugging the device in, but
> not all of the time.
> 
> The reason seems to be that usbmodules is getting called from the
> hotplug package at the same time the hub driver is talking to the
> device.  Since we don't have control endpoint locking in the kernel USB
> code yet, this messes things up :)
> 
> The patch below, from Olaf Hering, seems to fix the problem.  Also just
> removing the usbmodules binary seems to fix it for other people, but
> that's not really a good fix.
> 
> Let me know what you think about this.
> 
> thanks,
> 
> greg k-h
> 
> 
> Index: etc/hotplug/usb.agent
> =================================> RCS file: /cvsroot/linux-hotplug/admin/etc/hotplug/usb.agent,v
> retrieving revision 1.19
> diff -a -u -r1.19 usb.agent
> --- etc/hotplug/usb.agent	18 Jan 2002 03:10:42 -0000	1.19
> +++ etc/hotplug/usb.agent	11 Mar 2002 21:52:10 -0000
> @@ -297,6 +297,9 @@
>  case $ACTION in
>  
>  add)
> +    # Let the usb subsystem finish talking to the device, before we do
> +    sleep 3
> +
>      usb_convert_vars
>  
>      FOUNDúlse


That patch does indeed fix the problem with some devices, sometimes you
even need a sleep 5.

But since all that runs in background the bootscripts will continue to
run. And if gpm wants the /dev/input/mice node and the mousedriver is
not loaded ...
Or printer.o and rclpd, everything fails :)

The delay has to be added, either to the hotplug scripts or to the
usbdevfs core. But it will delay the driver loading in both cases, so
the startup scripts may fail. It depends on your configuration and the
speed of your cpu and/or IO.


Gruss Olaf

-- 
 $ man clone

BUGS
       Main feature not yet implemented...

_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: [PATCH] fix some hubs and hid devices at startup

[Greg KH]

On Wed, Mar 20, 2002 at 03:53:47PM +0100, Olaf Hering wrote:
> That patch does indeed fix the problem with some devices, sometimes you
> even need a sleep 5.
> 
> But since all that runs in background the bootscripts will continue to
> run. And if gpm wants the /dev/input/mice node and the mousedriver is
> not loaded ...
> Or printer.o and rclpd, everything fails :)
> 
> The delay has to be added, either to the hotplug scripts or to the
> usbdevfs core. But it will delay the driver loading in both cases, so
> the startup scripts may fail. It depends on your configuration and the
> speed of your cpu and/or IO.

Yes, it's a bandage over the real problem, sorry.

But here's a patch from Itai Nahshon that seems to fix a lot of problems
that he was having with the latest 2.4.19-pre3 USB code when a device is
connected to the bus.  It also seems to fix a lot of issues a few
keyboards and hubs that I was seeing here.

I've applied this to both the 2.4 and 2.5 trees.

thanks,

greg k-h


diff -Nru a/drivers/usb/hub.c b/drivers/usb/hub.c
--- a/drivers/usb/hub.c	Wed Mar 20 09:15:02 2002
+++ b/drivers/usb/hub.c	Wed Mar 20 09:15:02 2002
@@ -655,11 +655,11 @@
  * Not covered by the spec - but easy to deal with.
  *
  * This implementation uses 400ms minimum debounce timeout and checks
- * every 10ms for transient disconnects to restart the delay.
+ * every 100ms for transient disconnects to restart the delay.
  */
 
 #define HUB_DEBOUNCE_TIMEOUT	400
-#define HUB_DEBOUNCE_STEP	10
+#define HUB_DEBOUNCE_STEP	100
 
 /* return: -1 on error, 0 on success, 1 on disconnect.  */
 static int usb_hub_port_debounce(struct usb_device *hub, int port)

_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel