From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1Ha9s6-00082G-EZ for qemu-devel@nongnu.org; Sat, 07 Apr 2007 08:18:22 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1Ha9s4-0007zi-TH for qemu-devel@nongnu.org; Sat, 07 Apr 2007 08:18:22 -0400 Received: from [199.232.76.173] (helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1Ha9s4-0007zf-N6 for qemu-devel@nongnu.org; Sat, 07 Apr 2007 08:18:20 -0400 Received: from nz-out-0506.google.com ([64.233.162.236]) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1Ha9oT-0006Em-Bv for qemu-devel@nongnu.org; Sat, 07 Apr 2007 08:14:37 -0400 Received: by nz-out-0506.google.com with SMTP id i11so1730902nzi for ; Sat, 07 Apr 2007 05:14:36 -0700 (PDT) Message-ID: <83a4d4ca0704070514v5e430de8h2eefd100e5ca274e@mail.gmail.com> Date: Sat, 7 Apr 2007 14:14:36 +0200 From: "Eduardo Felipe" Subject: Re: [Qemu-devel] Feature proposal: USB devices over TCP In-Reply-To: <46171239.6090608@codemonkey.ws> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_18001_32853967.1175948076434" References: <83a4d4ca0704061811ka631da0ufad5af1a63ba19c2@mail.gmail.com> <46171239.6090608@codemonkey.ws> Reply-To: qemu-devel@nongnu.org List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org ------=_Part_18001_32853967.1175948076434 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi Anthony, I don't know all that much about USB, but I think the most useful way to > do this would be to do something that's protocol compatible with USBIP. Neither I do. I didn't know anything about USB/IP, it would be cool to be protocol compatible with it. As we have to start from scratch, USB/IP can be a very good reference. We could then tunnel this traffic over VNC and allow for exposing local > USB devices to a remote VM. I don't think that tunneling traffic through VNC is possible. In QEMU the VNC implementation is asynchronous, while the USB layer expects an immediate response to any request. Think of a virtual desktop being hosted on a server and exposed on a > thin client. If you could plug in your iPod and it would just work with > the VM, that would be an exceedingly cool feature. > > Are you familiar with USBIP? If so, does this sound reasonable? It would be cool indeed. Maybe some wrapper around libusb could do the trick for host OSes where no USB/IP server is available. I'll have a look at USB/IP. Regards, Eduardo Felipe ------=_Part_18001_32853967.1175948076434 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

Hi Anthony,

I don't know all that much about USB, but I think the most useful way to
do this would be to do something that's protocol compatible with USBIP.

Neither I do. I didn't know anything about USB/IP, it would be cool to be protocol compatible with it. As we have to start from scratch, USB/IP can be a very good reference.

We could then tunnel this traffic over VNC and allow for exposing local
USB devices to a remote VM.

I don't think that tunneling traffic through VNC is possible. In QEMU the VNC implementation is asynchronous, while the USB layer expects an immediate response to any request.

Think of a virtual desktop being hosted on a server and exposed on a
thin client.  If you could plug in your iPod and it would just work with
the VM, that would be an exceedingly cool feature.

Are you familiar with USBIP?  If so, does this sound reasonable?

It would be cool indeed. Maybe some wrapper around libusb could do the trick for host OSes where no USB/IP server is available.

I'll have a look at USB/IP.

Regards,
Eduardo Felipe
------=_Part_18001_32853967.1175948076434--