At 04:46 AM 10/7/2005 +0200, Eric Auer wrote:
The test already is performed inside the driver, and cannot be performed
in an errorlevel-generating tool. Because errorlevels are not used in
context of DEVICE=... loading. The "autoconfigure" would look like this:
Seriously, you think that stacking up duplicate drivers and expecting them
to all fail gracefully in line with the current CPU is a good way to handle
the default boot process? You want to depend on a solid but harmless fail
condition to make things work every time?
I'm not much of one to volunteer people into more kernel work, but for
cripes sake, if you're doing to do any conditional configurations at
boot-time on the CONFIG.SYS, _that_ is where it belongs. IF
<application/driver> and IFNOT <application/driver> conditionals would do
the trick. It's not like CONFIG.SYS processing hasn't already been
extended for FreeDOS a lot more than that already.
Or possibly easier...have the boot loader/new installation optionally able
to launch an appropriate CONFIG conditionally based on the detected
CPU. Start with a CONFIG.386 -- and for the benighted few a CONFIG.86 and
CONFIG.286 -- whichever one is correct gets brought in during installation
and renamed to everybody's favorite CONFIG.SYS at the end of install.
Just think, with conditional CONFIG.SYS processing, you could even expand
flexibility to dynamically bypass known problem -- or include required --
driver configurations with SCSI drives, flash drives, goofball BIOSs,
etc. Consequently you might become a world-wide marvel, millions possibly
singing hosannas to your name, and you could be more popular than Linus
(not Torvalds or Pauling, of course, I mean Linus Scudwomper who inflates
balloons and female dolls for a living in Trenton, New Jersey, but still...).
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
Freedos-user mailing list