On Tuesday 01 December 2009 2:20:32 pm Charles Owens wrote: > Charles Owens wrote: > > Howdy, > > > > I'm still digging into it, put preliminary testing is showing that > > with a PAE-enabled 7.1-RELEASE-p8 kernel a USB keyboard functions only > > occasionally, if you're lucky. This flakiness is being seen with > > several different server models that we support (HP and IBM, all Xeon > > based). If we boot a non-PAE kernel then the USB keyboard is rock solid. > > My initial report was entirely too broad --- an unrelated problem was > confusing things. I can now report that this problem appears limited to > only the IBM System x3550 M2. With a PAE-enabled kernel initialization > of some of the USB buses fails. From an external perspective, this > results in the rear USB port being non-functional, while the two front > USB ports work fine. For some reason, though, when booting the same PAE > kernel from a CD none of the USB ports function. > > In previous posting ( > http://docs.freebsd.org/cgi/mid.cgi?4B0D14CB.1010000 ) I included output > of usbdevs and boot log when running the PAE-enabled kernel. Note the > "root hub error" messages in boot log. > > Below I'll repeat the same info when running a non-PAE kernel. > > Any thoughts as to what the story is here... and suggestions for > resolution? Thanks in advance.
Would you be able to test an 8.0 kernel with PAE enabled? 8.0 has a completely rewritten USB stack, so there is a chance that the bug in 7 might not be present in 8. You can also try posting on freebsd-usb@ since the folks working on the USB drivers tend to hang out on that list. -- John Baldwin _______________________________________________ [email protected] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hardware To unsubscribe, send any mail to "[email protected]"
