On Thu, 25 Jul 2013, James Stone wrote: > On Thu, Jul 25, 2013 at 7:26 PM, Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> wrote: > > On Thu, 25 Jul 2013, James Stone wrote: > > > >> > Please try the patch from here: > >> > > >> > http://marc.info/?l=linux-usb&m=137476385206265&w=2 > >> > > >> > instead of the one I sent to you yesterday. I think it will fix this > >> > problem. > >> > > >> > Alan Stern > >> > > >> > >> Perfect! All fixed! > > > > Can you provide a usbmon trace showing the start of a session using the > > new patch? > > > > Alan Stern > > > > Sure - will send off list. The only slight difference I can see is > that maybe the 3.10 uses slightly higher CPU load than 3.5 at the > ridiculously low latency of 64 frames/period duplex. With the new patch, what you actually get is 44.1 frames/period (on average). That's quite sustainable. However, something's not working right. The number of packets in each playback URB changes each time the URB is reused! That's not supposed to happen. The number of packets should remain fixed while the number of samples in each packet changes, based on the feedback info. I don't get it. The usbmon trace shows three URBs, and the number of packets goes like this: 8 8 8 8 4 8 4 8 3 8 4 8 4 8 3 8 4 8 3 8 4 8 4 8 etc... I'm at a complete loss. It's not caused by confusion on the feedback endpoint; in fact the driver doesn't change this number anywhere once it has been set up. What's going on? James, please apply the diagnostic patch below on top of everything else. It will write entries into the kernel log which hopefully will pin down where this number gets changed. Another problem, not necessarily a bad one: The feedback data from the sound device indicates that its internal clock is actually running at 45168 Hz, even though it claims to be running at 44100. Alan Stern Index: usb-3.10/sound/usb/endpoint.c =================================================================== --- usb-3.10.orig/sound/usb/endpoint.c +++ usb-3.10/sound/usb/endpoint.c @@ -33,6 +33,8 @@ #include "pcm.h" #include "quirks.h" +static int alancnt = -1; + #define EP_FLAG_ACTIVATED 0 #define EP_FLAG_RUNNING 1 #define EP_FLAG_STOPPING 2 @@ -358,10 +360,16 @@ static void snd_complete_urb(struct urb goto exit_clear; if (usb_pipeout(ep->pipe)) { + if (alancnt > 0) + printk(KERN_INFO "B urb %p cnt %d\n", + urb, urb->number_of_packets); retire_outbound_urb(ep, ctx); /* can be stopped during retire callback */ if (unlikely(!test_bit(EP_FLAG_RUNNING, &ep->flags))) goto exit_clear; + if (alancnt > 0) + printk(KERN_INFO "C urb %p cnt %d\n", + urb, urb->number_of_packets); if (snd_usb_endpoint_implicit_feedback_sink(ep)) { unsigned long flags; @@ -375,6 +383,13 @@ static void snd_complete_urb(struct urb } prepare_outbound_urb(ep, ctx); + if (urb->number_of_packets != 8 && alancnt < 0) + alancnt = 20; + if (alancnt > 0) { + printk(KERN_INFO "A urb %p cnt %d\n", + urb, urb->number_of_packets); + --alancnt; + } } else { retire_inbound_urb(ep, ctx); /* can be stopped during retire callback */ -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
