On Thu, 4 Aug 2005, Erik Walthinsen wrote: > I'm building a multi-seat system, with 6 video cards intended to be all > run separately, so 6 people can sit down at the machine at the same > time. This for a lab to be installed in a school in the middle of > Bolivia next month. > > I've gotten the X issues worked through (but not entirely solved), but > now I'm running into USB issues. Specifically, I can get all the > keyboards to work just fine, they have never given me any problems at all. > > However, the mice are a whole other story. I've seen at most 3 mice > work at any given time, and plugging in a fourth will sometimes result > in the latest mouse working, sometimes not. Either way the total number > of working mice stays at 3 or below. > > Each keyboard and mouse pair is attached to a dedicated USB2 hub, with > each hub dedicated a port on the motherboard's root hub, conveniently > labeled 1..6 in the resulting USB device paths. > > Plugging in the three hubs, one at a time, with keyboard and mouse > already attached, generally results in the first two mice working and > the third not. Sometimes it's the first mouse that stops working and > the last one starts working. Rarely they all work. > > Starting with just the hubs, I can plug in the keyboards one at a time, > no problems. Then start adding mice, testing each one as I go, and I > get similar results, sometimes they all work, sometimes it's the first > or last mouse that stops working. > > I haven't been able to discern anything out of the ordinary from the > kernel logs, unfortunately. My testing as to whether or not the devices > work is a python script I wrote, using an 'evdev.py' out on the net, > which will open *every* /dev/input/event* and check for updates every > 100msec, printing out the new data if there are any changes. Some > -HUP's patched into the hotplug system *usually* inform my script of new > or removed devices, at which point it tosses the lot and re-opens > everything. > > Part of the puzzle is almost certainly that sometimes this -HUP doesn't > work, so I have to restart the script to see devices that indeed work. > there are no protections against nested -HUPs in my script, so that > makes me think that somewhere the hotplug scripts are screwing up when > there are two or more running in parallel, e.g. at bootup when > theoretically all 6 hubs, keyboards, and mice are all triggering hotplug > at the same time. > > For reference I'm using the latest stock Ubuntu 5.04 kernel, which > currently is linux-image-2.6.10-5-386 (2.6.10-34.3). I can try a newer > kernel as needed. > > If anyone has any insight into what on earth is going on, I would very > much appreciate it. I have a dog&pony show at 6pm PST Friday and I'd > love to have this solved by then, if possible.
You should try sending email to the people who maintain the USB HID driver or to the people in charge of the input subsystem. Those people generally don't read linux-usb-users. Alan Stern ------------------------------------------------------- SF.Net email is Sponsored by the Better Software Conference & EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf _______________________________________________ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
