I'm in trouble to get some usb(4) devices working here. Devices are

attached as ugen(4) and are driven by userspace (libusb) drivers/daemons.
As far as I can see what happens is that daemon opens ugen(4) device

correctly and starts to poll(2) interrupt endpoint of the device (devices

have single interface). The trouble is that poll(2) seems to ignore

timeout. It doesn't return until there is data available in the

descriptor. Ie I can make poll(2) to return with poking device, but till

then it blocks.
Before I dig deep into areas I'm not familiar with ... Any ideas why and

where it might happen? Anyone seen something like this?
--

Hasso Tepper

In which waitchan is the process?  If it is in "poll" and you're really

passing a proper timeout value, then everything should work as advertised.

  Does a ktrace show only the poll and no return?  Sometimes processes

tend to re-call poll over and over again (maybe due to an interrupting

signal).
cheers

   simon
--

Serve - BSD     +++  RENT this banner advert  +++    ASCII Ribbon   /"\

Work - Mac      +++  space for low €€€ NOW!1  +++      Campaign     \ /

Party Enjoy Relax   |   http://dragonflybsd.org      Against  HTML   \

Dude 2c 2 the max   !   http://golden-apple.biz       Mail + News   / \

Indeed ... Well, back to debug ugen(4), there is something very fishy

going on.
--

Hasso Tepper