Date: Wed, 21 Jan 2004 17:04:09 +1100 (EST) From: Bruce Evans <bde@zeta.org.au> To: Nate Lawson <nate@root.org> Cc: Kris Kennaway <kris@obsecurity.org> Subject: Re: cvs commit: src/sys/alpha/alpha support.s src/sys/i386/i386 swtch.s src/sys/kern kern_shutdown.c src/sys/sys systm.h Message-ID: <20040121165555.T7572@gamplex.bde.org> In-Reply-To: <20040120191020.J99547@root.org> References: <200401192127.i0JLRBL3041817@repoman.freebsd.org> <20040120015009.W39477@odysseus.silby.com> <20040121004238.GP47639@wantadilla.lemis.com> <20040121022454.GA28078@xor.obsecurity.org> <20040120191020.J99547@root.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 20 Jan 2004, Nate Lawson wrote: > > On Tue, Jan 20, 2004 at 05:04:12PM -0800, Nate Lawson wrote: > > > I would have gotten 100x more value from making options DDB the default in > > > GENERIC (at least until we branch -stable) and having all page faults > I just looked and GENERIC still has options DDB (and INVARIANTS, etc.) > So all we need is re@ to leave options DDB on for the release builds. It > adds little space and users can compile it out. But for "panic on > install" problems or novice users, this would be very useful. Also, > having re@ provide a kernel.debug for each GENERIC would be good too. It > could be on the live FS cd, not on cd1. DDB would need to be made more secure and robust for production use. It can be invoked by anyone at the console with a default syscons console and keymap. It cannot be invoked by a break or escape sequence on the default serial console, but that makes it less useful. The debug.debugger_on_panic sysctl doesn't work right. It only affects panics, but for production use all stray calls to Debugger() should cause panics and not hang waiting for input in ddb. Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040121165555.T7572>