From owner-freebsd-current Thu Oct 8 21:30:12 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA16277 for freebsd-current-outgoing; Thu, 8 Oct 1998 21:30:12 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from spinner.netplex.com.au (spinner.netplex.com.au [202.12.86.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA16178 for ; Thu, 8 Oct 1998 21:29:53 -0700 (PDT) (envelope-from peter@netplex.com.au) Received: from spinner.netplex.com.au (localhost [127.0.0.1]) by spinner.netplex.com.au (8.9.1/8.9.1/Spinner) with ESMTP id LAA04031; Fri, 9 Oct 1998 11:57:21 +0800 (WST) (envelope-from peter@spinner.netplex.com.au) Message-Id: <199810090357.LAA04031@spinner.netplex.com.au> X-Mailer: exmh version 2.0.2 2/24/98 To: Greg Lehey cc: Doug Rabson , Brian Feldman , Brian Feldman , current@FreeBSD.ORG Subject: Re: LKM conversions (was: your mail) In-reply-to: Your message of "Fri, 09 Oct 1998 10:42:02 +0930." <19981009104202.G3369@freebie.lemis.com> Date: Fri, 09 Oct 1998 11:57:20 +0800 From: Peter Wemm Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Greg Lehey wrote: > On Thursday, 8 October 1998 at 11:05:22 +0100, Doug Rabson wrote: > > On Wed, 7 Oct 1998, Brian Feldman wrote: > > > >> I meant, how does it _look_, because it seems to work just fine, despite > >> lower-level kldload things not working. such as, green_Saver kldloading > >> will cause a trap 12 in IP 0xf012929d, lookup_symbol I believe. No working > >> panics, so no good debugging :( Lemme guess, you can't panic on a DDB > >> trap? Brian, the code in the tree at present is horrifially incomplete and has some real problems. > > Most of the LKMs have not been converted to the new style. I believe that > > Peter Wemm is working on this right now. You might like to drop him a > > line. > > Vinum has not been (and will not be by 15 October) converted. There > are several reasons for this: > > 1. The Vinum LKM was committed two minutes before a deadline on 15 > September. > > 2. At this point, there had been no mention that LKMs were not > supported in 3.0, nor that the kernel would be ELF. > > 3. At this point, no LKMs had been converted to KLDs. > > 4. The current ELF kernel doesn't support ddb, so it's impractical to > develop code on this platform. Yes, it does. It just hasn't been committed yet until it's *clear* that it does and will all work. Even then, there will be absolutely nothing stopping somebody building an a.out kernel, or whatever. The stuff that's being worked on is not a forced change, it's meant to get the technology out there. I'm running right now (on the machine that I'm typing this) on an ELF kernel that supports dynamic modules, either loaded via kldload(8) or via /boot/loader. peter@overcee[12:12pm]~-100> kldstat Id Refs Address Size Name 1 2 0xf0100000 17f998 kernel 2 1 0xf0280000 1b20 testmod.ko Testmod.ko (like a .so, but it's *not* PIC so there's no speed penalty) was automatically "there" as a result of my boot.conf file. DDB can see all this stuff just nicely, *including* the testmod.ko symbols - no more tracebacks showing offsets from _end. I have a couple more things to fix and then it's done. BTW; LKM's convert to KLD modules quite simply, but the tools to do it are not in the tree. A modified gensetdefs is needed (at present). > Greg Cheers, -Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message