Bug 26061

Summary:	4.2 ata driver fails on CTX laptop
Product:	Base System	Reporter:	RonDzierwa <RonDzierwa>
Component:	kern	Assignee:	Søren Schmidt <sos>
Status:	Closed FIXED
Severity:	Affects Only Me
Priority:	Normal
Version:	Unspecified
Hardware:	Any
OS:	Any

Description RonDzierwa 2001-03-25 06:50:02 UTC

I have used this laptop with 2.2.8 for about two years.  I
recently upgraded the disk and decided to upgrade to a newer
version of FreeBSD as well.  The old wd driver worked fine
(my old 2.2.8 boot disk finds the new disk as well as the 
old one). 

Why does the ata driver think it has found ata2 and 3 instead
of ata0 and 1?

How-To-Repeat: boot the install flop

Comment 1 Kris Kennaway freebsd_committer

2001-03-26 01:41:24 UTC

Responsible Changed
From-To: freebsd-bugs->sos

sos is Mr ATA

Comment 2 Søren Schmidt freebsd_committer

2001-04-02 20:17:49 UTC

State Changed
From-To: open->Analyzed

I have no idea why this happens actually, so you need 
to sit down and try to debug why this fails, evt by 
spprinkling printf's around the probe/attach code. 
If you get me a prober description of what goes wrong 
I'm sure it can be fixed.

Comment 3 Søren Schmidt freebsd_committer

2001-05-30 08:43:17 UTC

State Changed
From-To: Analyzed->closed

The problem is that the CTX has bogus PCI reg values 
that screw up the probes, this is not easily worked 
around, and will probably break alot of other (correct) 
machines. My suggestion is to use a local hack for 
this box (which already is done).

Comment 4 RonDzierwa 2001-05-30 19:04:36 UTC

I am not using a local hack.  I am using the WD driver, which works.
It seems to me the ata driver IS a local hack that only works on the
small universe of machines that Soren has available to him.  His driver
sucks.  AND he sucks as well since he is unwilling to fix his bugs. I
have worked in software engineering for over 20 years, and no self
respecting engineer would let this kind of thing go by, especially when
he has been offered help in fixing it!  This ata driver is the only
piece
of software that cannot understand the ide controller on this laptop -
Windoze 95/98/NT all have no problems.  Even Linux 6.2 and 7.1
understand it.  In his last message, Soren said:

"Other than that it can only be solved by adding even more heuristics
to the pci probe stuff (which needs ALOT of testing to make sure it
doesn't break elsewhere)

-Søren

what he needs is less heuristics, and a little more smarts.  If he
really
did ALOT of testing in the first place, we would have never gotten
in this situation.  It seems to me he just got it working for a handful
of machines that he encountered, and never really designed it to
work with the real world.

This issue has become a black eye for FreeBSD, at least in the area
where I work.  It saddens me to see a hacker such as Soren make
the whole effort look bad to so many people.  RedHat, here I come.

ron.