Vesa fix

Vesa fix

[James Simmons]

Hi!
  
   The results of the EDID info from the BIOS has been varied. For some it 
worked. Others it gave the wrong data. Then for other even the headers 
where corrupted :-( So this patch pulls out the EDID code from Vesa. The 
EDID code works for PPC tho :-)

Re: Vesa fix

[James Simmons]

> 
> Hi!
>   
>    The results of the EDID info from the BIOS has been varied. For some it 
> worked. Others it gave the wrong data. Then for other even the headers 
> where corrupted :-( So this patch pulls out the EDID code from Vesa. The 
> EDID code works for PPC tho :-)
> 
> 
> -

Oop.s Forgot patch.


# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
#	           ChangeSet	1.1078  -> 1.1079 
#	drivers/video/vesafb.c	1.31    -> 1.32   
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 03/05/12	jsimmons@maxwell.earthlink.net	1.1079
# [VESA FBDEV] Removed the EDID code. The results where mixed. It worked for some but not for others.
# --------------------------------------------
#
diff -Nru a/drivers/video/vesafb.c b/drivers/video/vesafb.c
--- a/drivers/video/vesafb.c	Mon May 12 14:26:05 2003
+++ b/drivers/video/vesafb.c	Mon May 12 14:26:05 2003
@@ -215,7 +215,6 @@
 int __init vesafb_init(void)
 {
 	int video_cmap_len;
-	char *edid = 0;
 	int i;
 
 	if (screen_info.orig_video_isVGA != VIDEO_TYPE_VLFB)
@@ -300,18 +299,10 @@
 		ypan = 0;
 	}
 
-#ifdef __i386__
-	edid = get_EDID_from_BIOS(0);
-	if (edid)
-		parse_edid(edid, &vesafb_defined);	
-	else		
-#endif
-	{	
-		/* some dummy values for timing to make fbset happy */
-		vesafb_defined.pixclock     = 10000000 / vesafb_defined.xres * 1000 / vesafb_defined.yres;
-		vesafb_defined.left_margin  = (vesafb_defined.xres / 8) & 0xf8;
-		vesafb_defined.hsync_len    = (vesafb_defined.xres / 8) & 0xf8;
-	}
+	/* some dummy values for timing to make fbset happy */
+	vesafb_defined.pixclock     = 10000000 / vesafb_defined.xres * 1000 / vesafb_defined.yres;
+	vesafb_defined.left_margin  = (vesafb_defined.xres / 8) & 0xf8;
+	vesafb_defined.hsync_len    = (vesafb_defined.xres / 8) & 0xf8;
 	
 	if (vesafb_defined.bits_per_pixel > 8) {
 		vesafb_defined.red.offset    = screen_info.red_pos;

Re: Vesa fix

[Dave Jones]

On Mon, May 12, 2003 at 10:44:29PM +0100, James Simmons wrote:

 >    The results of the EDID info from the BIOS has been varied. For some it 
 > worked. Others it gave the wrong data. Then for other even the headers 
 > where corrupted :-( 

XFree86 seems to do a pretty good job of getting it right, maybe they
have blacklists ? Then again, this stuff is arguably better off done
in userspace anyway IMO.

		Dave

Re: Vesa fix

[Alan Cox]

On Maw, 2003-05-13 at 18:50, Dave Jones wrote:
> On Mon, May 12, 2003 at 10:44:29PM +0100, James Simmons wrote:
> 
>  >    The results of the EDID info from the BIOS has been varied. For some it 
>  > worked. Others it gave the wrong data. Then for other even the headers 
>  > where corrupted :-( 
> 
> XFree86 seems to do a pretty good job of getting it right, maybe they
> have blacklists ? Then again, this stuff is arguably better off done
> in userspace anyway IMO.

XFree handles EDID per driver and uses different methods in different
drivers.

Re: Vesa fix

[James Simmons]

> > XFree86 seems to do a pretty good job of getting it right, maybe they
> > have blacklists ? Then again, this stuff is arguably better off done
> > in userspace anyway IMO.

Not helpful on platforms which lack a hardware text mode.
 
> XFree handles EDID per driver and uses different methods in different
> drivers.

So I found out the hardware. BIOS are evil!!!!!