Re: [Qemu-devel] [PATCH] usb-redir: Allow to attach USB 2.0 devices to 1.1 host controller

[Hans de Goede <hdegoede@redhat.com>]

Hi,

On 09/21/2012 01:49 PM, Jan Kiszka wrote:
> On 2012-09-18 11:41, Hans de Goede wrote:
>> Hi,
>>
>> On 09/17/2012 06:22 PM, Jan Kiszka wrote:
>>>> If that does not work, add the debug parameter to the usb-redir device,
>>>> set it
>>>> to 4, collect logs of trying to redirect the device and send me the logs
>>>> please, ie:
>>>> -device usb-redir,chardev=usbredirchardev1,id=usbredirdev1,debug=4
>>>>
>>>> Also be aware that usb-redir relies on chardev flowcontrol working,
>>>> which it does not upstream! See for example here for the chardev flow
>>>> control patch set which RHEL / Fedora carry:
>>>> http://cgit.freedesktop.org/~jwrdegoede/qemu/log/?h=qemu-kvm-1.2-usbredir&ofs=50
>>>>
>>>>
>>>> And then the first 13 patches after: "Merge tag 'v1.2.0'"
>>>>
>>>> Oh, and also, if you're running qemu git master, make sure you've:
>>>> http://cgit.freedesktop.org/~jwrdegoede/qemu/commit/?id=81e34f5973d8d6a1ef998a50c4a4bf66abb3b56b
>>>>
>>>
>>> I used qemu-kvm-1.2-usbredir^ (the last commit is apparently broken -
>>> copy&paste bug?).
>>
>> Yeah, that has been fixed now.
>>
>>> I'm getting this right after typing cat /dev/ACM0 in
>>> the guest. It's an endless stream, and so is the output in the guest
>>> although there should be nothing to dump (that's the proper behaviour on
>>> the host).
>>
>> Hmm, can you try commenting out line 1608 of hw/usb/redirect.c:
>>               usb_ep->pipeline = true;
>>
>> And see if that helps. If it does not help, please bump the debug level to 5
>> (this will also make it log packet contents), and then generate another log, and
>> then it is time to dive into the ACM protocol to see what is happening...
>
> As it looks like now, I was just using the wrong test on the guest side.
> Retried this morning briefly with a terminal program, and it was all
> fine, even when forwarding from host-ehci to guest-uhci (with my broken
> patch), even when using current QEMU git head. Sorry for the noise

Ok, so to be clear: this is solved now, right ?

Regards,

Hans