Date: Sun, 31 Jul 2005 00:36:52 +0200 From: Oliver Lehmann <lehmann@ans-netz.de> To: current@freebsd.org Cc: imp@freebsd.org Subject: Re: mlx(4) fully and ep(4) partially broken on 6.0 Message-ID: <20050731003652.12223405.lehmann@ans-netz.de> In-Reply-To: <20050730014342.0843080d.lehmann@ans-netz.de> References: <20050730014342.0843080d.lehmann@ans-netz.de>
next in thread | previous in thread | raw e-mail | index | archive | help
Oliver Lehmann wrote: > 1st problem with 6.0: > --------------------- > > The system kept rebooting automaticly after it detected ep0. It was > saying sth. like "ep0: no irq?!". The funny thing is, one line above it > wrote sth. like "ep0 <3COM....> irq 9 ....". for me that looks like ep0 > has of course an irq assigned. > > I am happy that I have physical access to this system! Damn! > > I changed the perfectly-on-5.4 working 3com card with an other I had > on storage - panic is gone now. Some things about this: I can reproduce this on my testsystem when I put "device intpm" into my kernel config. intpm0 runs with irq 9: intpm0 <Intel 82371AB Power management controller> port 0x5000-0x500f ir1 9 at device 7.3 on pci0 same as ep0 wants to run with: ep0: <3Com 3C509-Combo Etherlink III> at port 0x300-0x30f irq 9 on isa0 ep0: No irq?! panic: resource_list_release: resource entry is not busy Uptime: 1s As you can see, it looks like the card is not configured as PnP (because that would add "(PnP)" to the description). On my router, fxp3 has irq 9 - Maybe it is irq sharing related? But I wonder why it works with 5.4 without problems... -- Oliver Lehmann http://www.pofo.de/ http://wishlist.ans-netz.de/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050731003652.12223405.lehmann>