Radeon 7500

[Stefan Bruda]

At 12:12 -0600 on 2002-9-7 Dan Burcaw wrote:
 >
 > > It sure doesn't sound like the video=radeon:dfp option worked at all.
 > > Am I doing something wrong here?
 > 
 > The BlueStone_B part just refers to your second video connector on that 
 > card I imagine.  Could you try a newer kernel (e.g. benh rsync) and
 > see if console switching works?

I have had a similar message (mine refers to "ATY,Crown_B," but the
idea is the same) for as long as I had my Radeon card, but I just
ignored it, as it is quite right: my guess is that the OF has room for
another card which is simply not there on my machine.  It does not
appear to refer to a video connector, these are detected a bit later
by the radeonfb kernel driver:

    radeonfb: DVI port LCD monitor connected
    radeonfb: CRT port no monitor connected

(the last line of course changes when I plug a monitor in the DVI
connector--the said monitor is correctly noticed).  Oh well, I am not
sure we refer to the same thing when we say "video connector," but in
any case the said "not found" message does not appear to have any
detrimental effect whatsoever.

In terms of settings, I boot with no video arguments (my day to day
boot arguments used to contain only "hdc=ide-scsi" since I use SCSI
emulation for my DVD/CD/CD-W drive, with the new benh kernels not even
this is necessary), which I believe is equivalent to OF video.  By the
way, I was under the impression that OF video just means that the
video settings are read from OF, the kernel driver fires up anyway, am
I right?

In X, I do use the "UseFBDev" option, in fact I have to: the X display
shows a legible image without it but it is not usable in the long run
(there are some lines and reflexions, which is of course no fun).  I
had to insert a new mode line for my display in XF86Config even if
XFree86 4.2 is said to be modeline-free.  In any case, X works well
with "UseFBDev."  The only problem I had is that sometimes the image
gets shaky after switching VTs or after (re)starting the X server.
But then I switch the VTs back and forth a couple of times (at most)
and this fixes the problem.

I am not sure how relevant this is for this discusion, as I am using
the (dreaded) M7 LW, aka 7500 Mobility (Dan, others please do
something about the sleep thing, pretty pplease ;-) ).  It is only
that the message the original poster refers to looks similar _and_
appears to be normal (in the sense that it makes sense and does not
appear to have any detrimental effect).

Oh, yeah, all of these are invariant with the kernel, provided that
the kernel is a reasonably new one.  In particular, I did not notice
any difference in this respect between the stock YDL 2.3 kernel
(2.4.19-4a) and the ones I have been using recently (benh rsync
2.4.19+, currently 2.4.20-pre5-ben0 or something).  The benh kernels
are custom build on my machine, but I guess it doesn't really matter
as the same thing happened with the 2.4.19-4a which I did not compile
myself.  (Incidentally though, benh kernels work better for me in a
number of other respects, so I do recommend them over the YDL kernel.)

Ron, I hope this helps, at least by eliminating some possible causes
for your problem.  Should you believe that I could be of further help
do not hesitate to get back to me.

Cheers,
Stefan

-- 
If it was so, it might be; and if it were so, it would be; but as
it isn't, it ain't. That's logic.
    --Lewis Carroll, Through the Looking-Glass