Date: Mon, 03 Nov 1997 20:09:56 -0800 From: Julian Elischer <julian@whistle.com> To: john hood <cgull+usenet-878428029@smoke.marlboro.vt.us> Cc: current@FreeBSD.ORG Subject: [try this] (Was: ddb broken in -current?) Message-ID: <345EA014.15FB7483@whistle.com> References: <Pine.BSF.3.95.971031003336.10876A-100000@current1.whistle.com> <199711020014.TAA09104@smoke.marlboro.vt.us>
next in thread | previous in thread | raw e-mail | index | archive | help
Anyone got a clue? 2 people have seen this now, can other people try this experiment? > Julian Elischer writes: > > > > boot a new -current kernel with: > > > > Boot: -sd > > > > when the kernel comes to a stop in the debugger add: > > > > break vfs_mountrootfs > > cont > > > > when the system breaks on vfs_mountrootfs() > > try > > > > trace > > > > or > > > > step (or in fact almost anything else) > > > > and get pagefaults.. > > > > Does any one else see this? > > (I updates my sources today) interestingly enough, 'cont' will allow you to continue. Though, since I'm trying to track down a crash this is not much help, because it immediatly hits my problem and crashes. john hood wrote: > > > It's a rather smart Heisenbug. > > I'm trying to track down some IDE problems (probably hardware), so I > try using the early-boot debugger, with 'boot -sd'. No matter what I > do in the debugger-- even just immediately continue out of it-- the > kernel fails to detect my IDE drives and panics on not finding a root > fs. If I set a breakpoint, after the breakpoint is hit, I get > pagefaults when I give any command. > > I think this bug knows where I live. > > I saw this about a week ago then got tied up by other things. This is also a kernel of about that vintage.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?345EA014.15FB7483>