Date: Wed, 3 Jan 1996 13:37:00 +1030 (CST) From: Michael Smith <msmith@atrad.adelaide.edu.au> To: jkh@time.cdrom.com (Jordan K. Hubbard) Cc: obrien@cs.ucdavis.edu, freebsd-hackers@freebsd.org Subject: Re: X for install Message-ID: <199601030307.NAA06495@genesis.atrad.adelaide.edu.au> In-Reply-To: <1164.820633374@time.cdrom.com> from "Jordan K. Hubbard" at Jan 2, 96 05:42:54 pm
next in thread | previous in thread | raw e-mail | index | archive | help
Jordan K. Hubbard stands accused of saying: > I've often stated my willingness to do the "GUI" grot that would be > required to make all those nifty auto-configuration screens come up, > but the `NTDETECT' side of things is not really in my area of > expertise. Knowing how to stomp around in the PC memory & I/O address > spaces and deal (hopefully) robustly with errors and cards going wiggy > when probed is a black art. Any black magicians out there interested > in starting a `FreeBSD Detect 1.0' project? :-) Sure. I require about half a million bucks to do it. The _only_ way to do this _properly_ is to buy several of everything. Microsoft can do this, we can't. 8( We _already_have_ detection code that handles lots of things pretty well, but our current boot-a-generic-kernel installation process is a _major_ bugbear; if we want to go poking and prodding hardware we want a _minimal_ kernel with just disk support and some way of getting a lumpier kernel onto the disk and set appropriately to match the hardware. This means a reboot during the install, or at least a reload of a new kernel. Unless we're really smart, it also means more than a single disk. How small can a kernel be made and still have all the disk drivers in it? We'd want all the SCSI disk drivers, wdc, sio, sc, UFS, CD9660 and a fixed 8M memory limit. No swap stuff, no networking, no quotas. Any DEVFS experts out there care to talk about how easy it is to start up device drivers after bootstrap now? If I have a device driver configured in and disabled, can I kick it later and have it probe for its hardware? This would allow for an alternative approach that has some merits; boot the kernel and then work through the disabled-configured devices in some experimentally-determined fashion, keeping notes along the way. > I can say one thing with absolute authority: The current "hardware > detection" scheme in sysinstall is utterly bogus, hateful and > genuinely evil. Agreed. > Jordan -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] "Who does BSD?" "We do Chucky, we do." [[
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199601030307.NAA06495>