Date: Sat, 25 Feb 2006 23:26:00 -0800 From: Garrett Cooper <youshi10@u.washington.edu> To: freebsd-questions@freebsd.org Subject: Re: anyone recognize this panic? Message-ID: <44015808.1000201@u.washington.edu> In-Reply-To: <20060226070300.GA59714@xor.obsecurity.org> References: <17409.8562.677322.222883@jerusalem.litteratus.org> <20060226033858.GA10985@xor.obsecurity.org> <440145EF.5000101@u.washington.edu> <20060226061343.GA4483@xor.obsecurity.org> <44014943.8050905@u.washington.edu> <20060226070300.GA59714@xor.obsecurity.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Kris Kennaway wrote: >On Sat, Feb 25, 2006 at 10:22:59PM -0800, Garrett Cooper wrote: > > > >>>>>>re0: diagnostic failed to receive packet in loopback mode >>>>>>re0: attach aborted due to hardware diagnostic failure >>>>>>panic: mtx_lock() of spin mutex (null) @ >>>>>>/usr/src/sys/netinet/in_pcb,c:862 >>>>>> >>>>>> "re0" is a Linksys EG-1032, less than two months old. It was >>>>>>connected, but had zero traffic at the time of the crash. >>>>>> Before I take this to current@ - has anyone seen anything like >>>>>>this before? A quick check of the archives and the web in general >>>>>>didn't show anything. >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>You need to at least get a traceback from the panic, and preferably a >>>>>crashdump. >>>>> >>>>>Kris >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>>Probably should pass this onto some devs. It seems like a null value was >>>>passed for locking a mutex in the OS, which is important. Having a >>>>traceback would be good though... but at least mentioning that the issue >>>>laid with the re (?) kernel driver would be a start. >>>> >>>> >>>> >>>> >>>Unfortunately without a traceback the panic string is useless since it >>>gives you no clue about how the system got into that state. This kind >>>of panic is often a secondary effect of some other problem. >>> >>>Kris >>> >>> >>> >>> >>True, but at least you'd be able to find a way to the affected code... >>After that it's just tests and debugging =\... >> >> > >I don't understand what you're suggesting; how do you find the >affected code without a traceback? > >Kris > I'm thinking of the "old fashioned way" of doing things... reading tons of code. Lol. -Garrett
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?44015808.1000201>