[PATCH] radeonfb ATI 9100IGP shared main memory / VRAM

[PATCH] radeonfb ATI 9100IGP shared main memory / VRAM

[Harald Welte]

[-- Attachment #1: Type: text/plain, Size: 2818 bytes --]

Hi Ben!

I have a couple of machines with an ASUS P4R800-VM mainboard.

lspci looks like this:

00:00.0 Host bridge: ATI Technologies Inc Radeon 9100 IGP Host Bridge (rev 02)
00:01.0 PCI bridge: ATI Technologies Inc Radeon 9100 IGP AGP Bridge
00:13.0 USB Controller: ATI Technologies Inc OHCI USB Controller #1 (rev 01)
00:13.1 USB Controller: ATI Technologies Inc OHCI USB Controller #2 (rev 01)
00:13.2 USB Controller: ATI Technologies Inc EHCI USB Controller (rev 01)
00:14.0 SMBus: ATI Technologies Inc ATI SMBus (rev 18)
00:14.1 IDE interface: ATI Technologies Inc: Unknown device 4349
00:14.3 ISA bridge: ATI Technologies Inc: Unknown device 434c
00:14.4 PCI bridge: ATI Technologies Inc: Unknown device 4342
00:14.5 Multimedia audio controller: ATI Technologies Inc IXP150 AC'97 Audio Controller
01:05.0 VGA compatible controller: ATI Technologies Inc Radeon 9100 IGP
02:08.0 Ethernet controller: 3Com Corporation 3Com 3C920B-EMB-WNM Integrated Fast Ethernet Controller (rev 40)
02:09.0 FireWire (IEEE 1394): Texas Instruments TSB43AB22/A IEEE-1394a-2000 Controller (PHY/Link)

The issue is that those boards (as I believe most 9100IGP chips do)
share main memory and video memory, and the user configures in CMOS
SETUP whihc amount of memory he wants to dedicate to the video card.

However (it might be a BIOS issue specific to this board), the cards
report 0k videoram.

radeonfb as it is in 2.6.14.4 therefore only does the following:

radeonfb_pci_register BEGIN
ACPI: PCI Interrupt 0000:01:05.0[A] -> GSI 16 (level, low) -> IRQ 177
radeonfb (0000:01:05.0): Found 0k of DDR 128 bits wide videoram
radeonfb (0000:01:05.0): cannot map FB
ACPI: PCI interrupt for device 0000:01:05.0 disabled
radeonfb: probe of 0000:01:05.0 failed with error -5

I've altered radeon_base.c with a special exception for this chip, see
the patch below.  Now everything works like a charm.

Cheers,
	Harald


--- linux-2.6.14.4/drivers/video/aty/radeon_base.c	2005-12-15 00:50:41.000000000 +0100
+++ linux-2.6.14.4-radeonfb/drivers/video/aty/radeon_base.c	2005-12-21 21:26:27.000000000 +0100
@@ -2157,6 +2157,11 @@
 		case PCI_CHIP_RADEON_LZ:
 	       		rinfo->video_ram = 8192 * 1024;
 	       		break;
+		case PCI_CHIP_RS300_5834:
+			/* This chip ?can? have shared main memory / VRAM,
+			 * and the size is configured according to CMOS SETUP */
+			rinfo->video_ram = pci_resource_len(rinfo->pdev, 0);
+			break;
 	       	default:
 	       		break;
 		}

-- 
- Harald Welte <laforge@gnumonks.org>          	        http://gnumonks.org/
============================================================================
"Privacy in residential applications is a desirable marketing option."
                                                  (ETSI EN 300 175-7 Ch. A6)

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [PATCH] radeonfb ATI 9100IGP shared main memory / VRAM

[Harald Welte]

[-- Attachment #1: Type: text/plain, Size: 725 bytes --]

On Wed, Dec 21, 2005 at 09:37:06PM +0100, Harald Welte wrote:
> I've altered radeon_base.c with a special exception for this chip, see
> the patch below.  Now everything works like a charm.

unfortunately I now discovered that the PCI resource is always 64MB in
size, even when the UMA BIOS setting is configured to smaller (e.g. 8MB)
sizes.

I don't have a clue for a proper fix, but maybe you have ;)

-- 
- Harald Welte <laforge@gnumonks.org>          	        http://gnumonks.org/
============================================================================
"Privacy in residential applications is a desirable marketing option."
                                                  (ETSI EN 300 175-7 Ch. A6)

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [PATCH] radeonfb ATI 9100IGP shared main memory / VRAM

[Benjamin Herrenschmidt]

On Wed, 2005-12-21 at 21:37 +0100, Harald Welte wrote:

> I've altered radeon_base.c with a special exception for this chip, see
> the patch below.  Now everything works like a charm.

Hi !

Actually, this isn't the right way, the PCI resource might be bigger
than the RAM assigned to the IGP chip. I'll try to get something better
for you. The thing to do looks like that in X:

    else if ((info->ChipFamily == CHIP_FAMILY_RS100) ||
	     (info->ChipFamily == CHIP_FAMILY_RS200) ||
	     (info->ChipFamily == CHIP_FAMILY_RS300)) {
        CARD32 tom = INREG(RADEON_NB_TOM);

	pScrn->videoRam = (((tom >> 16) -
			    (tom & 0xffff) + 1) << 6);

	OUTREG(RADEON_CONFIG_MEMSIZE, pScrn->videoRam * 1024);

With
	#define RADEON_NB_TOM                       0x15c

Ben.




-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click

Re: Re: [PATCH] radeonfb ATI 9100IGP shared main memory / VRAM

[Benjamin Herrenschmidt]

On Wed, 2005-12-21 at 23:26 +0100, Harald Welte wrote:
> On Wed, Dec 21, 2005 at 09:37:06PM +0100, Harald Welte wrote:
> > I've altered radeon_base.c with a special exception for this chip, see
> > the patch below.  Now everything works like a charm.
> 
> unfortunately I now discovered that the PCI resource is always 64MB in
> size, even when the UMA BIOS setting is configured to smaller (e.g. 8MB)
> sizes.
> 
> I don't have a clue for a proper fix, but maybe you have ;)

You didn't get my mail about using RADEON_NB_TOM instead ?

Ben.




-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click