AMD A8 3850 APU "Heart Transplant" - Experiences

English Hardware
#1

I am 4 hours into a heart transplant for my 11.4/KDE 4.7.1 System.

I replaced an Asus M2AVM-HDMI/QuadCore/8G DDR2/XFX PCI video platform with an
Asus F1A75M-Pro/A8 3850 APU/16G DDR3 platform, planning to use the onboard graphics.

I had earlier migrated** / **to a 64GB SSD, /home to a 1.5T hdd and a 2T hdd /backup drive, all SATAIII.

After swapping out the Motherboard and connecting up the SSD/HDDs. I encountered 3 issues coming back on line:

  1. The new F1A75 MoBo was in serious need of a BIOS upgrade; I used the EZ update built-in(reads new file from USB stick).
    No complaint here, this is relatively new mobo
  2. Network did not come up - so at level 3 I ran YAST and configured the new Eth0 device(same chip as before, but new MAC id)
  3. Getting X and fglrx running took a bit - finally realized that xorg.conf was the problem.
    Running aticonfig – initial located the APU graphics section, I then edited the new PCI address (BusID “PCI:0:1:0”) into my old xorg.conf and I was back up and running with dual monitor (side by side) configuration.

An anomaly for PTA:

When I tried atiupgrade, I received a message “No ATI Device found, proceed[yn]?”.
I said “y”, the process proceeded and completed.
I am curious as to why the script didn’t find the ATI 9640 “Device”, which of course is on-chip in the APU.
Need more diagnostic info?

The PCI info follows

lspci
00:00.0 Host bridge: Advanced Micro Devices [AMD] Device 1705
00:01.0 VGA compatible controller: ATI Technologies Inc Device 9640
00:01.1 Audio device: ATI Technologies Inc Device 1714
00:10.0 USB Controller: Advanced Micro Devices [AMD] Hudson USB XHCI Controller (rev 03)
00:10.1 USB Controller: Advanced Micro Devices [AMD] Hudson USB XHCI Controller (rev 03)
00:11.0 SATA controller: Advanced Micro Devices [AMD] Hudson SATA Controller [IDE mode] (rev 40)
00:12.0 USB Controller: Advanced Micro Devices [AMD] Hudson USB OHCI Controller (rev 11)
00:12.2 USB Controller: Advanced Micro Devices [AMD] Hudson USB EHCI Controller (rev 11)
00:13.0 USB Controller: Advanced Micro Devices [AMD] Hudson USB OHCI Controller (rev 11)
00:13.2 USB Controller: Advanced Micro Devices [AMD] Hudson USB EHCI Controller (rev 11)
00:14.0 SMBus: Advanced Micro Devices [AMD] Hudson SMBus Controller (rev 13)
00:14.1 IDE interface: Advanced Micro Devices [AMD] Hudson IDE Controller
00:14.2 Audio device: Advanced Micro Devices [AMD] Hudson Azalia Controller (rev 01)
00:14.3 ISA bridge: Advanced Micro Devices [AMD] Hudson LPC Bridge (rev 11)
00:14.4 PCI bridge: Advanced Micro Devices [AMD] Hudson PCI Bridge (rev 40)
00:14.5 USB Controller: Advanced Micro Devices [AMD] Hudson USB OHCI Controller (rev 11)
00:15.0 PCI bridge: Advanced Micro Devices [AMD] Device 43a0
00:15.1 PCI bridge: Advanced Micro Devices [AMD] Device 43a1
00:15.2 PCI bridge: Advanced Micro Devices [AMD] Device 43a2
00:18.0 Host bridge: Advanced Micro Devices [AMD] Family 12h/14h Processor Function 0 (rev 43)
00:18.1 Host bridge: Advanced Micro Devices [AMD] Family 12h/14h Processor Function 1
00:18.2 Host bridge: Advanced Micro Devices [AMD] Family 12h/14h Processor Function 2
00:18.3 Host bridge: Advanced Micro Devices [AMD] Family 12h/14h Processor Function 3
00:18.4 Host bridge: Advanced Micro Devices [AMD] Family 12h/14h Processor Function 4
00:18.5 Host bridge: Advanced Micro Devices [AMD] Family 12h/14h Processor Function 6
00:18.6 Host bridge: Advanced Micro Devices [AMD] Family 12h/14h Processor Function 5
00:18.7 Host bridge: Advanced Micro Devices [AMD] Family 12h/14h Processor Function 7
03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 06)
04:00.0 USB Controller: Device 1b21:1042

I have not done significant testing as yet, but it sure does boot fast!

#2

Here is an update on my issue with the atiupgrade script:

Looking at the script, the following line of code finds the ati model

ati_model=$(lspci | sed -n '/VGA/s/.*Radeon *//p' | tr -d "]" )

The A8 3550 based system says:

lspci
00:01.0 VGA compatible controller: ATI Technologies Inc Device 9640

I note that my HPDV7 laptop with chipset embedded graphics says:

lspci
01:05.0 VGA compatible controller: ATI Technologies Inc RS780M/RS780MN [Radeon HD 3200 Graphics]

The issue is that the script expects the “Radeon xxxx Graphics” string (for this device xxxx = 6550) but it is not reported by lspci.

I did allow atiupgrade to proceed, by answering “y” to the query, and the latest driver was downloaded, compiled and installed.

Later, however, I did notice some intermittent loading of OpenGL and some error messages in /var/log/Xorg.0.log.
The cure was to exit to a console (Ctl-Alt F1), log in a root, then

init 3
rmmod fglrx
init 5

The new fglrx module loaded and OpneGL up and running with no errors.

#3

Yes, you got it. You can see that the script is looking for the string “Radeon” and didn’t find it. If it would look for ATI, it would find older chipsets as well, that are not Radeon. The main purpose of this check is to avoid installing the driver on chipsets where it wouldn’t work (that would be worse), but it you know what you want and what you are doing, you can allow the script to proceed, just as you did. In fact, I do sometimes download and package the driver on nvidia systems.
What does hwinfo report in your case?

hwinfo --gfxcard | grep -i -e model

Possibly newer chipsets are not “Radeon” anymore (?).

#4

This will not be much help

hwinfo --gfxcard | grep -i -e model
  Model: "ATI VGA compatible controller"

This from a Tom’s Hardware item a while back:

Several reports are indicating that AMD’s upcoming line of Llano Fusion accelerated (APUs) will feature the Radeon HD 6550 graphics core.

I assume the string delivered to lspci comes from microcode in the chip, so unless AMD standardizes what it programs in I guess we’ll have to manually work around it.

#5

I just updated the script: Upgrading ATI driver with atiupgrade - Page 9

Next time you run it, it will skip the model detection if it finds an fglrx package or module (so it will assume that you have an ATI card supported by the driver). If you run the latest version of atiupgrade (2.9), it should just tell you that the latest Catalyst version is already installed*.

  • today.
#6

I had done some fighting during the weekend with the same APU. The script work A-O-K on 11.4. I have to say that no one will protest if similar functions where standard in the future release’s of Opensuse. I’m running the lates 12.1 Beta+factory on that PC and I get working anyway with Catalyst 11.9 after some work around.

Tanks anyway.