Date: Wed, 21 Mar 2007 08:49:38 +1100 From: Nick Withers <nick@nickwithers.com> To: Jean-Yves Lefort <jylefort@FreeBSD.org> Cc: freebsd-gnome@freebsd.org, manphiz@gmail.com Subject: Re: Possible missing file in hal installation Message-ID: <20070321084938.8d98861f.nick@nickwithers.com> In-Reply-To: <20070320181822.c02595b9.jylefort@FreeBSD.org> References: <20070319205346.e5e46409.nick@nickwithers.com> <20070320063524.421ad658.jylefort@FreeBSD.org> <20070320213838.dbef0f41.nick@nickwithers.com> <20070320181822.c02595b9.jylefort@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 20 Mar 2007 18:18:22 +0100 Jean-Yves Lefort <jylefort@FreeBSD.org> wrote: > On Tue, 20 Mar 2007 21:38:38 +1100 > Nick Withers <nick@nickwithers.com> wrote: > > > On Tue, 20 Mar 2007 06:35:24 +0100 > > Jean-Yves Lefort <jylefort@FreeBSD.org> wrote: > > > > > On Mon, 19 Mar 2007 20:53:46 +1100 > > > Nick Withers <nick@nickwithers.com> wrote: > > > > > > > Firstly, I've just pasted the existing thread in from > > > > "http://lists.freebsd.org/pipermail/freebsd-gnome/2007-January/016501.html", > > > > sorry if I stuffed it up at all and my apologies to those with > > > > threaded mail clients... > > > > > > > > > On Wed, 2007-01-03 at 21:13 +0800, manphiz at gmail.com > > > > > wrote: > > > > > > Joe Marcus Clarke wrote: > > > > > > > No, there is nothing useful here. The ktrace looks > > > > > > > good. The last thing hald does is read the PCI config > > > > > > > registers (part of the initial PCI bus probe). You > > > > > > > need to get a full backtrace from hald once it appears > > > > > > > to lock up. To do that, you should run hald manually, > > > > > > > then use gdb to bind to it once it's locked up. Then > > > > > > > get a full backtrace. > > > > > > > > > > > > > > Joe > > > > > > > > > > > > > > > > > > > > Thanks a lot for the instruction! I attached to the > > > > > > deadlocked process, and get this backtrace: > > > > > > > > > > > > (gdb) bt > > > > > > #0 hf_pci_probe () at hf-pci.c:150 > > > > > > #1 0x0805ced7 in osspec_probe () at osspec.c:96 > > > > > > #2 0x080551dd in main (argc=3, argv=0xbfbfeb5c) at > > > > > > #hald.c:578 > > > > > > > > > > This backtrace doesn't make sense. Line 150 of hf-pci.c is > > > > > not in hf_pci_probe(). The real line 150 represents a loop > > > > > that iterates through the devices list. This list must be > > > > > finite. I don't see how this problem could occur based on > > > > > this information. > > > > > > > > I appear to be in a similar scenario on my HP Compaq Presario > > > > B3820 (believe I'm using the latest firmware versions). > > > > > > > > (snip) > > > > > > If anyone's got an idea or two up their sleeves about how I > > > > might go about fixing this, I'd be very grateful to hear > > > > about it! > > > > > > Please install sysutils/pciutils and send me the output of > > > "lspci -v". > > > > With pleasure! > > ____ > > > > presario# lspci -v > > pcilib: Bus 02 seen twice (firmware bug). Ignored. (snip) > > ____ > The attached patch should fix the problem. Please test it. Worked an absolute charm (and how quick was that?!) - thanks very much Jean-Yves, very much appreciated! > -- > Jean-Yves Lefort > > jylefort@FreeBSD.org > http://lefort.be.eu.org/ -- Nick Withers email: nick@nickwithers.com Web: http://www.nickwithers.com Mobile: +61 414 397 446
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070321084938.8d98861f.nick>