From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:57747) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TugRn-0005SE-7A for qemu-devel@nongnu.org; Mon, 14 Jan 2013 04:35:21 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1TugRh-0007lw-0O for qemu-devel@nongnu.org; Mon, 14 Jan 2013 04:35:15 -0500 Received: from isrv.corpit.ru ([86.62.121.231]:53077) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TugRg-0007Z8-P5 for qemu-devel@nongnu.org; Mon, 14 Jan 2013 04:35:08 -0500 Message-ID: <50F3D145.3050903@msgid.tls.msk.ru> Date: Mon, 14 Jan 2013 13:35:01 +0400 From: Michael Tokarev MIME-Version: 1.0 References: <20120807012056.11172.7440.malonedeb@gac.canonical.com> <20130110215918.18012.96269.malone@chaenomeles.canonical.com> <50F11CB8.90800@msgid.tls.msk.ru> <50F3CD7D.1080907@redhat.com> In-Reply-To: <50F3CD7D.1080907@redhat.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [Bug 1033727] Re: USB passthrough doesn't work anymore with qemu-kvm 1.1.1 List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Gerd Hoffmann Cc: qemu-devel@nongnu.org 14.01.2013 13:18, Gerd Hoffmann =D0=BF=D0=B8=D1=88=D0=B5=D1=82: > On 01/12/13 09:20, Michael Tokarev wrote: >> Gerd, can you take a (maybe quick) look please? >> >> This bug is quite old and, as Peter notes, may be fixed in 1.3 >> so there's no goo motivation to fix it for older version... >> I tried to reproduce (to be able to get a trace, or bisect) it >> several times on my machine but can't, even with devices other >> people reported as triggering it. Please give some advice for >> Debian users ;) > > Given the amount of fixes which went into usb (not just usb-host, but > also uhci/ehic emulation) since 1.1 there is nothing but upgrading I ca= n > seriously suggest ... Well. Big share of these fixes are in the new async code which were added for 1.2. There aren't really much fixes for the original wrong behavour. I were looking at all usb-related changes since 1.1 up to current 1.3, and there aren't really many fixes in there which are not already picked for -stable. But ofcourse I overlooked the actual fix for this problem, which appears to be this commit: commit 72a04d0c178f01908d74539230d9de64ffc6da19 Author: Hans de Goede Date: Wed Sep 12 15:08:40 2012 +0200 uhci: Don't queue up packets after one with the SPD flag set Don't queue up packets after a packet with the SPD (short packet de= tect) flag set. Since we won't know if the packet will actually be short = until it has completed, and if it is short we should stop the queue. This fixes a miniature photoframe emulating a USB cdrom with the wi= ndows software for it not working. Signed-off-by: Hans de Goede Signed-off-by: Gerd Hoffmann which I wanted to queue for 1.1-stable too, initially, but decided it isn't really needed ;) Thank you! /mjt