On Wed, 28 Mar 2012, Chen Peter-B29397 wrote:
> I have no experiences on controller without build-in TT. According to your words,
> When a full/low speed device attaches, without build-in TT controller's PE will still
> be 0 even reset completes. So:
That's right. The port can't be enabled if the device doesn't send a
high-speed "chirp".
> ehci_dbg (ehci, "port %d full speed --> companion\n",
> index + 1);
> may be useful for them.
>
> How about only change the code like below:
>
> diff --git a/drivers/usb/host/ehci-hub.c b/drivers/usb/host/ehci-hub.c
> index 77bbb23..7e35e59 100644
> --- a/drivers/usb/host/ehci-hub.c
> +++ b/drivers/usb/host/ehci-hub.c
> @@ -536,7 +536,9 @@ static int check_reset_complete (
> if (ehci->has_amcc_usb23)
> set_ohci_hcfs(ehci, 1);
> } else {
> - ehci_dbg (ehci, "port %d high speed\n", index + 1);
> + if (!ehci_is_TDI(ehci)) {
> + ehci_dbg (ehci, "port %d high speed\n", index + 1);
> + }
> /* ensure 440EPx ohci controller state is suspended */
> if (ehci->has_amcc_usb23)
> set_ohci_hcfs(ehci, 0);
But this doesn't log any message if there is a builtin TT!
Look, your original message made a good point. We don't need to see
the speed here because khubd will display it later on. All we care
about is that the reset was complete, and whether or not the port got
enabled.
If you want, just replace the existing message with
ehci_dbg(ehci, "port %d reset complete, port enabled\n",
index + 1);
Alan Stern
P.S.: Putting "stable@xxxxxxxxxx" in the CC: list of the email message
is _not_ the right way to submit a patch for the stable kernel series.
Instead, you should put "CC: <stable@xxxxxxxxxxxxxxx> following your
"Signed-off-by:" line.
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
