From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?utf-8?B?TWF0xJtq?= Laitl Subject: Re: Allocating buffers for USB transfers (again) Date: Fri, 12 Aug 2011 01:29:53 +0200 Message-ID: <3166003.Y8gKYdUrh9@edgy> References: <20110811180507.GC5884@xanatos> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Daniel Mack Cc: Sarah Sharp , Alan Stern , Florian Mickler , Oliver Neukum , linux-usb@vger.kernel.org, alsa-devel@alsa-project.org, Takashi Iwai , Clemens Ladisch , pedrib@gmail.com, William Light , Greg KH , linux-kernel@vger.kernel.org, Robert Hancock List-Id: alsa-devel@alsa-project.org On 11.=E2=80=AF8.=E2=80=AF2011 Daniel Mack wrote: > On Thu, Aug 11, 2011 at 8:05 PM, Sarah Sharp > > Because according to Matej, he applied that patch, plus my patch to > > reject zero-length buffers[1], and he saw debugging that indicated = he > > *did* see zero-length buffers. Is there any chance your driver mig= ht > > submit a zero-length buffer in the middle of the isochronous URB > > transfer array? >=20 > Hmm, judging from the code, this can only ever happen if we receive a= n > inbound iso frame which has a valid status and an actual_length of > zero. Also, it was not neccessary to catch this case for EHCI. >=20 > Maetj, does this patch make any difference? This patch actually makes the sound playback _worse_. Now I get strange= =20 squawks where previously at least first seconds of a song sounded norma= lly. However, I no longer get "zero length buffer submitted" or that "... We= ird."=20 debug messages, only several megabytes of: (should I post these somewhe= re?) xhci_hcd 0000:05:00.0: Giveback URB ffff880114cec000, len =3D 880, expe= cted =3D=20 1000, status =3D -115 xhci_hcd 0000:05:00.0: Giveback URB ffff880114740000, len =3D 352, expe= cted =3D=20 1000, status =3D -115 xhci_hcd 0000:05:00.0: underrun event on endpoint Regards, Matej