Greg, ping?
On Thu, Nov 08, 2012 at 01:57:04PM -0800, Sarah Sharp wrote:
> The math in the link status reporting is wrong. The link state is in
> bits 8:5 of the hub port status word, which is split across a char
> array. Currently, if the link state is 0xb (Loopback mode), the status
> char array will look something like this:
>
> status[0] = 0110 0000
> status[1] = 0000 0001
>
> The current math will mask and shift those bits like so:
>
> (status[0] & 0xe0) >> 5 = 0000 0011
> (status[1] & 0x1) << 4 = 0001 0000
>
> This sets link_state to 0x13, which is not correct (it should be 0xb).
> This bug will cause valid link states greater than seven (Recovery, Hot
> Reset, Compliance Mode, and Loopback Mode) to not be printed.
>
> It's important to correctly report Compliance Mode, since some TI USB
> 3.0 host controllers can get "stuck" in Compliance Mode.
>
> Fix this by left shifting status[1] by 3, not 4.
>
> Signed-off-by: Sarah Sharp <sarah.a.sharp@xxxxxxxxxxxxxxx>
> ---
> lsusb.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/lsusb.c b/lsusb.c
> index 106dd33..5f0df11 100644
> --- a/lsusb.c
> +++ b/lsusb.c
> @@ -3276,7 +3276,7 @@ static void do_hub(libusb_device_handle *fd, unsigned tt_type, unsigned speed)
> (status[0] & 0x01) ? " connect" : "");
> } else {
> link_state = ((status[0] & 0xe0) >> 5) +
> - ((status[1] & 0x1) << 4);
> + ((status[1] & 0x1) << 3);
> printf("%s%s%s%s%s%s",
> (status[2] & 0x80) ? " C_CONFIG_ERROR" : "",
> (status[2] & 0x40) ? " C_LINK_STATE" : "",
> --
> 1.7.9
>
> --
> 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
--
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
