Blade 100, 2.6.13.2, atyfb

Blade 100, 2.6.13.2, atyfb

[Kumba]

[-- Attachment #1: Type: text/plain, Size: 746 bytes --]


Couple minor issues noted on Blade 100 running a 2.6.13.2 kernel.  They're 
probably already known, but...

A) atyfb seems h0rked still.  I think this has been looked at, but I don't 
recall a resolution yet.

B) atyfb's build seems broken due to the following:

   LD      vmlinux
drivers/built-in.o: In function `atyfb_xl_init':
: undefined reference to `aty_st_lcd'
make: *** [vmlinux] Error 1

The attached patch (attempts to) correct for this based on some studying of the 
function causing the problem.


--Kumba

-- 
Gentoo/MIPS Team Lead
Gentoo Foundation Board of Trustees

"Such is oft the course of deeds that move the wheels of the world: small hands 
do them because they must, while the eyes of the great are elsewhere."  --Elrond

[-- Attachment #2: sparc64-atyfb_xl_init.patch --]
[-- Type: text/plain, Size: 494 bytes --]

--- drivers/video/aty/xlinit.c.orig	2005-09-18 21:50:18.000000000 -0400
+++ drivers/video/aty/xlinit.c	2005-09-18 21:50:28.000000000 -0400
@@ -252,9 +252,11 @@ int atyfb_xl_init(struct fb_info *info)
 	aty_st_le32(0xEC, 0x00000000, par);
 	aty_st_le32(0xFC, 0x00000000, par);
 
+#ifdef CONFIG_FB_ATY_GENERIC_LCD
 	for (i=0; i<sizeof(lcd_tbl)/sizeof(lcd_tbl_t); i++) {
 		aty_st_lcd(lcd_tbl[i].lcd_reg, lcd_tbl[i].val, par);
 	}
+#endif
 
 	aty_st_le16(CONFIG_STAT0, 0x00A4, par);
 	mdelay(10);

Re: Blade 100, 2.6.13.2, atyfb

[Tom 'spot' Callaway]

[-- Attachment #1: Type: text/plain, Size: 1180 bytes --]

On Sun, 2005-09-18 at 22:35 -0400, Kumba wrote:
> Couple minor issues noted on Blade 100 running a 2.6.13.2 kernel.  They're 
> probably already known, but...
> 
> A) atyfb seems h0rked still.  I think this has been looked at, but I don't 
> recall a resolution yet.

I can confirm this as well.

> B) atyfb's build seems broken due to the following:
> 
>    LD      vmlinux
> drivers/built-in.o: In function `atyfb_xl_init':
> : undefined reference to `aty_st_lcd'
> make: *** [vmlinux] Error 1
> 
> The attached patch (attempts to) correct for this based on some studying of the 
> function causing the problem.

Aurora's had a similar patch in our 2.6 kernel for some time now. The
only difference is that we move the int i; variable declaration inside
the conditional to avoid unnecessary compiler noise. 

(I thought I had sent this to the list long ago, but I didn't. My bad.)

Signed-off-by: Tom 'spot' Callaway <tcallawa@redhat.com>

~spot
-- 
Tom "spot" Callaway: Red Hat Senior Sales Engineer || GPG ID: 93054260
Fedora Extras Steering Committee Member (RPM Standards and Practices)
Aurora Linux Project Leader: http://auroralinux.org
Lemurs, llamas, and sparcs, oh my!

[-- Attachment #2: linux-2.6.11-atyfb-nolcd.patch --]
[-- Type: text/x-patch, Size: 809 bytes --]

--- linux-2.6.11/drivers/video/aty/xlinit.c.BAD	2005-03-04 20:01:29.526690531 -0500
+++ linux-2.6.11/drivers/video/aty/xlinit.c	2005-03-04 20:04:49.384307555 -0500
@@ -174,7 +174,7 @@ int atyfb_xl_init(struct fb_info *info)
 	const struct xl_card_cfg_t * card = &card_cfg[xl_card];
 	struct atyfb_par *par = (struct atyfb_par *) info->par;
 	union aty_pll pll;
-	int i, err;
+	int err;
 	u32 temp;
 	
 	aty_st_8(CONFIG_STAT0, 0x85, par);
@@ -252,9 +252,12 @@ int atyfb_xl_init(struct fb_info *info)
 	aty_st_le32(0xEC, 0x00000000, par);
 	aty_st_le32(0xFC, 0x00000000, par);
 
+#if defined (CONFIG_FB_ATY_GENERIC_LCD)
+	int i;
 	for (i=0; i<sizeof(lcd_tbl)/sizeof(lcd_tbl_t); i++) {
 		aty_st_lcd(lcd_tbl[i].lcd_reg, lcd_tbl[i].val, par);
 	}
+#endif
 
 	aty_st_le16(CONFIG_STAT0, 0x00A4, par);
 	mdelay(10);

Re: Blade 100, 2.6.13.2, atyfb

[David S. Miller]

From: "Tom 'spot' Callaway" <tcallawa@redhat.com>
Date: Sun, 18 Sep 2005 23:18:47 -0500

> Signed-off-by: Tom 'spot' Callaway <tcallawa@redhat.com>

Applied, thanks Tom.