Re: vesa(4) and amd64

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

b. f. wrote:
> After the recent x86emu/vesa/dpms commits, I'm now able to use some
> more graphics modes with syscons on amd64. That's good.  Not so good
> is the fact that my HP Pavilion desktop running 9-CURRENT i386 r197085
> with
> 
> devic sc
> options SC_PIXEL_MODE
> device vga
> options VGA_WIDTH90
> 
> in the kernel and agp, dpms, x86emu, and vesa loaded as kernel modules
> can no longer use the 132x60 mode that had been my default syscons
> mode, and now yields a blank screen.  Even worse is the fact that my
> Toshiba laptop, with nearly the same configuration, locks up with a
> screen full of zeroes every time I load the new vesa kernel module,
> when formerly it had no such problem.  Other than simplifying the
> organization of the code, is there any advantage to be gained from
> forcing those platforms that are capable of native vesa to use x86emu?
>  Because up to this point there are serious disadvantages to doing so.

I think it was caused by some unrelated change.  ddkprog_at_ has proposed a
change, here is a slightly modified one, could you please give it a try?
 I'll try to see if I can have some clue myself tonight.

Cheers,
- --
Xin LI <delphij_at_delphij.net>	http://www.delphij.net/
FreeBSD - The Power to Serve!
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.12 (FreeBSD)

iEYEARECAAYFAkqrAXwACgkQi+vbBBjt66AK3wCgj9fnz60SWIIa7OUAdF/4x8aR
evsAoJ3A8QObHWMYXsOXwKbuCBR0pxKe
=Sr3u
-----END PGP SIGNATURE-----

Index: sys/dev/fb/vesa.c
===================================================================
--- sys/dev/fb/vesa.c	(revision 197050)
+++ sys/dev/fb/vesa.c	(working copy)
_at__at_ -1126,7 +1126,7 _at__at_
 	} else {
 		vesa_adp->va_buffer = 0;
 		vesa_adp->va_buffer_size = info.vi_buffer_size;
-		vesa_adp->va_window = (vm_offset_t)(emumem+L_ADD(info.vi_window));
+		vesa_adp->va_window = info.vi_window + KERNBASE;
 		vesa_adp->va_window_size = info.vi_window_size;
 		vesa_adp->va_window_gran = info.vi_window_gran;
 	}