USB config SX lock deadlock

2013-10-09 Thread Daniel O'Connor
Hi, We have a custom Cypress FX2 based data acquisition card connected to a FreeBSD 9.2 system. Today the device disconnected Oct 9 03:24:11 juskym kernel: ugsio0: at uhub3, port 1, addr 3 (disconnected) and then some time later we noticed the data acquisition process was stuck in 'USB config

Re: USB config SX lock deadlock

2013-10-09 Thread Hans Petter Selasky
On 10/09/13 12:04, Daniel O'Connor wrote: Hi, We have a custom Cypress FX2 based data acquisition card connected to a FreeBSD 9.2 system. Today the device disconnected Oct 9 03:24:11 juskym kernel: ugsio0: at uhub3, port 1, addr 3 (disconnected) and then some time later we noticed the data ac

Re: USB config SX lock deadlock

2013-10-09 Thread Daniel O'Connor
On 09/10/2013, at 20:59, Hans Petter Selasky wrote: >> I have previously run the same hardware on 9.0 (where there was a panic in >> devfs triggered if the DAQ card disconncted sometimes) and 9.1 (where I have >> not seen any problems). >> >> Does anyone have an idea about how I could go about

Re: USB config SX lock deadlock

2013-10-09 Thread Hans Petter Selasky
On 10/09/13 12:38, Daniel O'Connor wrote: On 09/10/2013, at 20:59, Hans Petter Selasky wrote: I have previously run the same hardware on 9.0 (where there was a panic in devfs triggered if the DAQ card disconncted sometimes) and 9.1 (where I have not seen any problems). Does anyone have an i

Re: USB device configuration

2013-10-09 Thread aseem.jolly
I couldn't really get the meaning of the following "Anyhow, if a second configure event fails after a successful one, we are still in the configured state". Are we making the following change done: if (err) unconfigure(); else usb_set_device_state(udev, USB_STATE_CO

Re: USB config SX lock deadlock

2013-10-09 Thread Daniel O'Connor
On 09/10/2013, at 22:43, Hans Petter Selasky wrote: > On 10/09/13 12:38, Daniel O'Connor wrote: >> On 09/10/2013, at 20:59, Hans Petter Selasky wrote: >> It is actually running r253505, sorry I should have mentioned that before. >> >> It isn't a serial device, although the driver is a thin wrap

Using a C-Media multichannel USB DAC

2013-10-09 Thread Bertrand Petit
Here on my desktop lies a new cheap multichannel brand-less USB DAC. I suppose it is based on a C-Media CM6206 chip. [1] From what I read in the datasheet [2] signals must properly be routed using HID requests before anything goes out of the connectors. How could this be done? My use case

Re: USB device configuration

2013-10-09 Thread Hans Petter Selasky
Hi, On 10/09/13 20:11, aseem.jolly wrote: I couldn't really get the meaning of the following "Anyhow, if a second configure event fails after a successful one, we are still in the configured state". It is allowed to set the configuration multiple times. Refer to USB 2.0 specification. Are

Re: Using a C-Media multichannel USB DAC

2013-10-09 Thread Hans Petter Selasky
On 10/10/13 05:13, Bertrand Petit wrote: Here on my desktop lies a new cheap multichannel brand-less USB DAC. I suppose it is based on a C-Media CM6206 chip. [1] From what I read in the datasheet [2] signals must properly be routed using HID requests before anything goes out of the conne

Re: USB config SX lock deadlock

2013-10-09 Thread Hans Petter Selasky
On 10/09/13 23:25, Daniel O'Connor wrote: On 09/10/2013, at 22:43, Hans Petter Selasky wrote: On 10/09/13 12:38, Daniel O'Connor wrote: On 09/10/2013, at 20:59, Hans Petter Selasky wrote: It is actually running r253505, sorry I should have mentioned that before. It isn't a serial device, al