linux-media.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* Magewell Gen 2935:0001 USB annoyances
@ 2019-01-14  5:10 Kai Hendry
  2019-01-14 11:30 ` Mauro Carvalho Chehab
  0 siblings, 1 reply; 3+ messages in thread
From: Kai Hendry @ 2019-01-14  5:10 UTC (permalink / raw)
  To: linux-media

Archlinux user here. It doesn't matter whether I'm running LTS kernel 4.19.14-1-lts or 4.20.1.arch1-1, I get these very annoying USB issues with my Magewell XI100DUSB-HDMI. Most of the time it doesn't work. I seemingly have better chances of getting it working after a reboot.

I don't know if this is a UVC issue or a USB issue on my Gen8 Thinkpad T480s. All I can say for a fact is that plugging this device into my Macbook Pro running MacOS via a USB-C dongle, is more reliable. I've had the same "plug in" issues admittedly on previous Thinkpad hardware and kernel versions.

Another annoying aspect, is that it also seems to drop USB speeds when hot plugging. I need USB 3 else it won't be able to get 60fps at 1080p.
https://s.natalian.org/2019-01-14/1547438681_2548x1398.png

It's a crazy ritual with accompanying dance when I get USB3 and my /dev/video2 actually working.

This is what dmesg looks like when it's flailing.

[   71.896534] usb 2-2: Disable of device-initiated U1 failed.
[   71.900554] usb 2-2: Disable of device-initiated U2 failed.
[   71.909474] usb 2-2: Set SEL for device-initiated U1 failed.
[   72.691371] usb 2-2: USB disconnect, device number 6
[   73.041273] usb 1-2: new high-speed USB device number 10 using xhci_hcd
[   73.620177] usb 1-2: New USB device found, idVendor=2935, idProduct=0001, bcdDevice= 0.00
[   73.620180] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[   73.620182] usb 1-2: Product: XI100DUSB-HDMI
[   73.620183] usb 1-2: Manufacturer: Magewell
[   73.620185] usb 1-2: SerialNumber: C021150326019
[   76.920993] usb 2-2: Set SEL for device-initiated U2 failed.
[   76.923341] uvcvideo: Found UVC 1.00 device XI100DUSB-HDMI (2935:0001)
[   76.925324] uvcvideo 1-2:1.0: Entity type for entity Processing 2 was not initialized!
[   76.925327] uvcvideo 1-2:1.0: Entity type for entity Camera 1 was not initialized!
[   76.925413] input: XI100DUSB-HDMI: XI100DUSB-HDMI  as /devices/pci0000:00/0000:00:14.0/usb1/1-2/1-2:1.0/input/input25
[   76.928710] usbhid 1-2:1.4: couldn't find an input interrupt endpoint
[   91.516482] usb 1-2: USB disconnect, device number 10


Is there any workarounds or ideas to make my capture device less annoying? Thank you in advance,

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Magewell Gen 2935:0001 USB annoyances
  2019-01-14  5:10 Magewell Gen 2935:0001 USB annoyances Kai Hendry
@ 2019-01-14 11:30 ` Mauro Carvalho Chehab
  2019-01-15  1:56   ` Kai Hendry
  0 siblings, 1 reply; 3+ messages in thread
From: Mauro Carvalho Chehab @ 2019-01-14 11:30 UTC (permalink / raw)
  To: Kai Hendry; +Cc: linux-media, Laurent Pinchart, linux-usb

Em Mon, 14 Jan 2019 13:10:25 +0800
Kai Hendry <hendry@iki.fi> escreveu:

> Archlinux user here. It doesn't matter whether I'm running LTS kernel 4.19.14-1-lts or 4.20.1.arch1-1, I get these very annoying USB issues with my Magewell XI100DUSB-HDMI. Most of the time it doesn't work. I seemingly have better chances of getting it working after a reboot.
> 
> I don't know if this is a UVC issue or a USB issue on my Gen8 Thinkpad T480s. All I can say for a fact is that plugging this device into my Macbook Pro running MacOS via a USB-C dongle, is more reliable. I've had the same "plug in" issues admittedly on previous Thinkpad hardware and kernel versions.
> 
> Another annoying aspect, is that it also seems to drop USB speeds when hot plugging. I need USB 3 else it won't be able to get 60fps at 1080p.

(adding c/c to linux-usb)

If the image is not compressed, a 640x480 image with 16 bits per pixel
and 30 frames per second eats 60% of the USB 2.0 maximum number of
ISOC frames with em28xx driver[1].

[1] The actual value is driver/device specific, as it depends on the
maximum frame size.

So, I'm pretty sure that you won't be able to get 60fps at 1080p
if it doesn't enable USB 3.0 SuperSpeed mode.

That's said, it is up to the USB stack to detect the device as an
USB 3.0. If this is not working well, it could either be a hardware
problem or some issue at the USB host driver.

> https://s.natalian.org/2019-01-14/1547438681_2548x1398.png
> 
> It's a crazy ritual with accompanying dance when I get USB3 and my /dev/video2 actually working.
> 
> This is what dmesg looks like when it's flailing.
> 
> [   71.896534] usb 2-2: Disable of device-initiated U1 failed.
> [   71.900554] usb 2-2: Disable of device-initiated U2 failed.
> [   71.909474] usb 2-2: Set SEL for device-initiated U1 failed.
> [   72.691371] usb 2-2: USB disconnect, device number 6
> [   73.041273] usb 1-2: new high-speed USB device number 10 using xhci_hcd
> [   73.620177] usb 1-2: New USB device found, idVendor=2935, idProduct=0001, bcdDevice= 0.00
> [   73.620180] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
> [   73.620182] usb 1-2: Product: XI100DUSB-HDMI
> [   73.620183] usb 1-2: Manufacturer: Magewell
> [   73.620185] usb 1-2: SerialNumber: C021150326019
> [   76.920993] usb 2-2: Set SEL for device-initiated U2 failed.

Everything here happens at USB host driver and USB subsystem.

> [   76.923341] uvcvideo: Found UVC 1.00 device XI100DUSB-HDMI (2935:0001)
> [   76.925324] uvcvideo 1-2:1.0: Entity type for entity Processing 2 was not initialized!
> [   76.925327] uvcvideo 1-2:1.0: Entity type for entity Camera 1 was not initialized!

The above is at uvcvideo, and seems OK to my eyes.

> [   76.925413] input: XI100DUSB-HDMI: XI100DUSB-HDMI  as /devices/pci0000:00/0000:00:14.0/usb1/1-2/1-2:1.0/input/input25
> [   76.928710] usbhid 1-2:1.4: couldn't find an input interrupt endpoint

This is related to input/evdev probing. Also unrelated to media.

> [   91.516482] usb 1-2: USB disconnect, device number 10

This is also at USB subsystem. Usually, it indicates a bad contact at the
USB cable.

> Is there any workarounds or ideas to make my capture device less annoying? Thank you in advance,

I would try to buy a high quality USB 3.0 cable, and see if things improve.

You should remember that a USB 3.0 data transfer will use higher
frequency signals, making the cable more susceptible to interference.

So, a cable with a good shield can make a difference, specially if
you're passing it close to other cables.

Thanks,
Mauro

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Magewell Gen 2935:0001 USB annoyances
  2019-01-14 11:30 ` Mauro Carvalho Chehab
@ 2019-01-15  1:56   ` Kai Hendry
  0 siblings, 0 replies; 3+ messages in thread
From: Kai Hendry @ 2019-01-15  1:56 UTC (permalink / raw)
  To: Mauro Carvalho Chehab; +Cc: linux-media, Laurent Pinchart, linux-usb

Thanks for getting back to me Mauro.

So it works on MacOS.... via a hootoo USB-C dongle https://s.natalian.org/2019-01-15/hootoo.jpeg

So since my T480s has two USB-C ports, I tried using the same dongle on my Thinkpad. It works!

I discovered that I can reliably get the device working by using another USB-C adaptor. https://s.natalian.org/2019-01-15/ss.jpeg

This to my astonishment reliably works.

[hendry@t480s ~]$ lsusb -vvv 2>/dev/null | grep -A5 2935:0001
Bus 004 Device 002: ID 2935:0001
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               3.00
  bDeviceClass          239 Miscellaneous Device


Maybe it's some USB bus power supply issue on my other Thinkpad ports. No idea.

Thanks for sounding me out.

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2019-01-15  1:56 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-01-14  5:10 Magewell Gen 2935:0001 USB annoyances Kai Hendry
2019-01-14 11:30 ` Mauro Carvalho Chehab
2019-01-15  1:56   ` Kai Hendry

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).