27-Jun-00 14:29 you wrote:
>>>>>> Khimenko Victor writes:
>>>> Ok. Just don't expect it from GRUB :-)
>>> Don't tell a lie.
KV>> "LIE" ?? It WAS truth and it IS truth.
> There is no need for anybody to get upset.
> Khimenko, it would have been much easier if you just said ``GRUB can
> do what you want using the `map' command, but it is inefficient and
> only works under DOS/Windows. To swap drives efficiently under all
> OSes cannot be done in general.''
This is basically what I've said... But if you want politcorrectenness...
Sorry - you come to wrong place.
>>>> I mean; stage2 can be on ide0). Read documentation about
>>>> installation process carefully.
>>> I'd like to say the same thing to you.
KV>> Hmm. Perhaps I've read it not carefully enough but I was unable
KV>> to find way to swap drives WITHOUT adding additional int13
KV>> handler (read: works with real OS) and without screwing
KV>> Windows9X. Am I missed something ?
> That constraint was not part of the original poster's question, so I
> fail to see why you are so adamant about it.
Original question was basically "I need to swap drives around in BIOS.
Is it possible to load this task on GRUB's shoulders ?". Obviusly the
goal is NOT to boot Windows from GRUB somehow. Goal was to start Windows
and WORK in it (otherwise why you started it at all?) but to not play
with BIOS settings to do so. And "map" command is bad helper here.
> He asked a simple question, and I supplied him with the only feasible answer.
FEASIBLE ? You gave answer that will "solve" his problem but then half year
later he'll find out that if he loading Windows from GRUB it works few times
slower then when Windows is loaded straight from SCSI disk with help of BIOS
swap. And what will be blamed for this slowness ? GRUB, of course - what else ?
User usually do not care (and do not know!) about INT13, BIOS acccess or
special driver access - it's not his work. He just want HIS work to be done.
Thus "without GRUB all works just fine, with GRUB Windows is slow like a
turtle => GRUB is piece of shit". It SUCH conclusion is what you want ? Hmm.
> Okuji was miffed at you because you were telling everybody that there
> was *no* solution,
And yes, there are NO sane solution to this "swap" problem. There IS sane
solution for ORIGINAL problem: just always use (scsi ide0 ide1) order and
teach GRUB about it.
> when in reality the only reasonable solution,
Reasonable ??? Hmm. Hardly.
> and the one GRUB implements, just didn't fit your own (unstated) constraint.
It's just simple Linus's rule:
-- cut --
Finally, even if the above isn't true, I often reject bug-fixes.
Bug-fixes are _often_ worse than the bug they fix, even with serious bugs.
Because a lot of bug-fixes are "band-aid" - not fixing the bug properly.
And band-aid is BAD. It's worse than even a crashing machine. Because
band-aid never goes away, and nobody cleans it up.
I prefer to have a known bug that will eventually get fixed than an ugly
solution that will hide it forever.
-- cut --
And "map" (when used with other system then DOS - even with Windows 3.11!)
is exactly such band-aid ugly solution - NOT proper fix.
> Is it clearer now why we saw your post as misinformed? Next time you
> post something that disregards another person's opinion, it is better
> to state your assumptions rather than sarcastically criticize their
> approach.
Hmm. I thought it's so obvious... When there is SANE and PROPER solution for
problem (not for directly asked question but for global problem) it's just
unwise to clog victim's mind with unneded (for him) information. Just state
that he chosen wrong way in solving bigger problem and thus smaller problem
he faced just do not have good solution (even if it HAS solution - it's
possible to cut tonsils from rear end with autogenous but do we REALLY need
to if more straight approach exist?).
