Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 18 Mar 2010 12:47:46 -0400
From:      Jung-uk Kim <jkim@FreeBSD.org>
To:        "barbara" <barbara.xxx1975@libero.it>
Cc:        freebsd-bugs <freebsd-bugs@FreeBSD.org>, dummy <dummy@smtp.ru>
Subject:   Re: kern/144654: [vesa] VESA support broken on FreeBSD 8.0 Stable [regression]
Message-ID:  <201003181247.58168.jkim@FreeBSD.org>
In-Reply-To: <KZEH5F$9E38F5F1B80E06837A974E1A48AC5922@libero.it>
References:  <KZEH5F$9E38F5F1B80E06837A974E1A48AC5922@libero.it>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday 16 March 2010 08:24 pm, barbara wrote:
> > On Tuesday 16 March 2010 06:32 pm, barbara wrote:
> > > > On Monday 15 March 2010 08:19 pm, barbara wrote:
> > > > > > On Monday 15 March 2010 02:41 pm, Jung-uk Kim wrote:
> > > > > > > Can you please try the attached patch?
> > > > > >
> > > > > > Oops, it attached a wrong patch.  Please try this
> > > > > > instead.
> > > > > >
> > > > > > Sorry for the inconvenience,
> > > > > >
> > > > > > Jung-uk Kim
> > > > >
> > > > > I had the same problem on RELENG_8 since yesterday (prev.
> > > > > buildworld on Feb. 28). I've tried rebuilding the kernel
> > > > > with your patch but no luck.
> > > >
> > > > Your problem may be different.  Different VESA BIOS has
> > > > different quirks.  Your BIOS actually sets non-VGA compatible
> > > > bits correctly. However, there is no standard VGA graphic
> > > > mode at all, which is pretty strange.
> > >
> > > Just for my personal understanding, can you explain me what
> > > does it means that "there is no standard VGA graphic mode at
> > > all" and what make you say that?
> >
> > Your "vidcontrol -i mode" output from the previous e-mail, i.e.,
> > all graphic mode belong to VESA mode (starting from 0x100).
>
> Thanks for explaining!
>
> > Are you sure you have SC_PIXEL_MODE in your kernel configuration?
>
> Yes, as you can seen from my reply to the other question.
>
> > > > What's your graphics card?
> > >
> > > vgapci0@pci0:2:0:0:     class=0x030000 card=0x00000000
> > > chip=0x042110de rev=0xa1 hdr=0x00 vendor     = 'Nvidia Corp'
> > >     device     = 'GeForce 8500 GT (G86)'
> > >     class      = display
> > >     subclass   = VGA
> >
> > Thanks.  I guess I have to buy one of those nVidia thingies. :-(
> >
> > > > > I had to comment the 'allscreens_flags="MODE_280"' line in
> > > > > my /etc/rc.conf as the monitor was going black with 'NO
> > > > > SIGNAL' on OSD. You asked to the OP if the box is pingable.
> > > > > I think that you want to know if the kernel is still alive,
> > > > > am I right? I can "blindly" login and reboot the pc while
> > > > > the screen is black.
> > > >
> > > > So, it is not rendering anything on screen.  Hmm...  Can you
> > > > please try the attached patch?
> > >
> > > No luck also with the new patch.
> > > I've applied it to the original file, not to the one with your
> > > previous patch, is that correct?
> >
> > Either way is fine.
> >
> > > It's not rendering anything because the led of the monitor
> > > turns from blue to orange, just like when the pc is turned off
> > > and the monitor is on. This happens running vidcontrol -i
> > > MODE_X where X is one of the modes of type "G" (as I've said I
> > > had to comment the allscreens_flags line in /etc/rc.conf). And
> > > if I subsequently switch to another ttyv with alt+f[1-8], the
> > > mode is not the default 80x25 as normally expected and the led
> > > is still orange and sometimes the speaker plays a beep!
> >
> > Okay, something is very wrong.  Can you please try the following?
> >
> > sysctl debug.x86bios.call=1
> > sysctl debug.x86bios.int=1
> > vidcontrol MODE_X
> > vidcontrol MODE_24	<- Sorry, you will have to blindly type it.
> >
> > And please send me the dmesg output.
>
> I did that with a script, so no blindly typing was necessary.
> I used MODE_280, which is the one I normally use.

:-)

> I see the following two lines appended to dmesg:
> Calling int 0x10 (ax=0x4f02 bx=0x4118 cx=0x0000 dx=0x0000 es=0x0000
> di=0x0000) Exiting int 0x10 (ax=0xc470 bx=0x0000 cx=0x52e4
> dx=0x0000 es=0xc000 di=0x0000) Is the output of the test ok? I mean
> is it what did you expected in terms of quantity?

No, I didn't expect it.  It seems the real mode emulator has 
terminated execution abruptly.  We need to gather more information 
off the list.

Jung-uk Kim



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201003181247.58168.jkim>