From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:40787) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TCXgb-0000qd-K0 for qemu-devel@nongnu.org; Fri, 14 Sep 2012 11:20:11 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1TCXgU-0005Lk-TR for qemu-devel@nongnu.org; Fri, 14 Sep 2012 11:20:04 -0400 Received: from mx1.redhat.com ([209.132.183.28]:54172) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TCXgU-0005LV-Ki for qemu-devel@nongnu.org; Fri, 14 Sep 2012 11:19:58 -0400 Message-ID: <50534B67.7030101@redhat.com> Date: Fri, 14 Sep 2012 17:21:11 +0200 From: Hans de Goede MIME-Version: 1.0 References: <1347525600-28220-1-git-send-email-kraxel@redhat.com> <5052E3DA.6000003@msgid.tls.msk.ru> <5052EA7F.6090105@redhat.com> In-Reply-To: <5052EA7F.6090105@redhat.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PULL 00/14] usb patch queue List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Gerd Hoffmann Cc: Michael Tokarev , qemu-devel@nongnu.org Hi, On 09/14/2012 10:27 AM, Gerd Hoffmann wrote: >> Are any of the bugfixes worth to apply to previous or new -stable >> series? I know usbredir configure fix is needed, anything else? > >>> Gerd Hoffmann (2): >>> usb-host: allow emulated (non-async) control requests without USBPacket > > This. > >>> ehci: Fix interrupts stopping when Interrupt Threshold Control is 8 > > This. > >>> usb-redir: Revert usb-redir part of commit 93bfef4c >>> uhci: Don't queue up packets after one with the SPD flag set > > And these two. > > I think the other ehci bugfixes are mostly needed for usbredir > migration, so I'd leave them out for stable. Hans? I would consider these stable candidates too: 1) ehci: Don't set seen to 0 when removing unseen queue-heads 2) ehci: Don't process too much frames in 1 timer tick (v2) 3) ehci: Fix interrupts stopping when Interrupt Threshold Control is 8 1) Fixes us canceling packets we should not cancel when we hit a high timer latency directly after the doorbell 2) Fixes streaming isoc data failing with "failed to resubmit urb (-27) errors in dmesg in the guest with Linux guests, this bug has existed for ages, but migration triggered it almost always, making me look into it 3) Is an obvious bugfix in itself, I'm not sure there are any guests hitting this, but the fix is so trivial it cannot hurt. Regards, Hans