On Mon, May 7, 2012 at 8:41 PM, Sarah Sharp <sarah.a.sharp@xxxxxxxxxxxxxxx> wrote: > On Fri, May 04, 2012 at 06:02:42PM -0400, Sergio Correia wrote: >> Hello folks, > > Hi Sergio! I'm the xHCI (USB 3.0 host controller) driver maintainer. > It helps if you Cc me directly on USB 3.0 issues; otherwise it ends up > in a mail folder I check less often. > Hi Sarah, and thanks for noticing this thread! My bad, I should have added you to CC from the beginning. The info you requested generated rather big files, so I hosted them somewhere else instead of attaching to this message. >> when I use the microphone of my Logitech headset plugged into a USB >> 3.0 port I get a robotic-like voice; if I plug it into a USB 2.0 port, >> it works normally. > > Can you send the output of `sudo lspci -vvv -n`? I'd like to know which > USB 3.0 host controller you have. > there: http://www.uece.net/misc/usb-headset-problem/lspci.vvv.n.txt >> I am testing on 3.4.0-rc5 with >> 0a6ba092d1c8c75ae0c0617ef7f4e6f5043af311 as the last commit [Merge tag >> 'mfd-for-linus-3.4-rc6' of >> git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-2.6]. >> I observe the same behavior with a 3.3.4 kernel. > > Has the USB 2.0 headset ever not had a robotic voice under your USB 3.0 > host? > It hasn't. I tried recently older kernels (3.2, 3.1 and 3.0) and they all exhibited this same behavior. The test succeeds on Windows, though; if I connect the headset to my USB 3.0 host it works like the 2.0 one. >> The device is this one: >> Bus 002 Device 004: ID 046d:0a0c Logitech, Inc. Clear Chat Comfort USB Headset > > Please run `sudo lsusb -vvv -d 046d:0a0c` with your device plugged into > the USB 3.0 port. I'm wondering if the bInterval reported in the > endpoint descriptors is incorrect, and you're getting a different > sampling rate under xHCI than EHCI. there: http://www.uece.net/misc/usb-headset-problem/lsusb.vvv.d.046d.0a0c.usb3.txt > >> Here are two (noisy) small audio samples to illustrate the difference >> between the recording if I use the USB 2.0 and 3.0 ports (no, I don't >> use this mic to record music, but I thought it would illustrate the >> problem heh) >> USB 2.0-> http://www.uece.net/misc/usb-headset-problem/test-usb2.0.mp3 >> USB 3.0-> http://www.uece.net/misc/usb-headset-problem/test-usb3.0.mp3 > > Yes, that does sound strange. :) > >> I am pasting my full dmesg and config below (did I miss any important >> stuff there?). > > Can you turn on CONFIG_USB_XHCI_HCD_DEBUGGING and CONFIG_USB_DEBUG, and > recapture dmesg when you plug in the device and start playing audio? It > may be a large log file, so you might need to put it somewhere or zip it > up. > Here's the dmesg without the device connected and when I have it plugged and capturing audio: http://www.uece.net/misc/usb-headset-problem/dmesg-no-device-connected-xhci-port.txt http://www.uece.net/misc/usb-headset-problem/dmesg-device-connected-xhci-port-recording.txt >> Let me know if is there any othe r useful info I could provide. > > It might help to compare the USBmon output from both EHCI and xHCI, when > the device is playing. Then we could see which frames the audio was > being transferred in. There is documentation on USBmon in the kernel > source tree: > > http://www.kernel.org/doc/Documentation/usb/usbmon.txt > Thanks for this; I am gonna try USBmon later on tonight or tomorrow and share the info. > Sarah Sharp thanks for your help! Sergio -- 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
