128GB BIOS for MS-5156 (TX4) needed

BIOS update, EIDE card, or overlay software? (FAQ Hard disk recognition)
hanker
New visitors - please read the rules.
Posts: 12
Joined: Tue Feb 13, 2007 6:49 pm

Hello!

I have an older Super Socket 7 - Mainboard from MSI, and need a 64GB or even better 128GB patch for my BIOS.

The Mainboard is a MS-5156 (TX4)-Mainboard you can find some more info at

http://www.msi-computer.de/produkte/mai ... Seite=BIOS (german site with beta BIOS)

or

http://www.msi.com.tw/program/support/d ... 215&kind=1 (english site with older BIOS and manual)

Can you help me? Do you need further info?

René
edwin
The Hardware Archivist
Posts: 6286
Joined: Wed Mar 20, 2002 7:11 pm
Location: Netherlands
Contact:

you've got mail
edwin/evasive

Do not assume anything

System error, strike any user to continue...
hanker
New visitors - please read the rules.
Posts: 12
Joined: Tue Feb 13, 2007 6:49 pm

Wow, that was fast!

I will flash the BIOS today, but currently have no big HDD (I willl buy it in some weeks). But I will tell you, if the PC still works and I recognize any problems.


Thanks again!

René
hanker
New visitors - please read the rules.
Posts: 12
Joined: Tue Feb 13, 2007 6:49 pm

Hi!

Well, I have had mixed feelings with the update and the whole "BIOS story".

First, the good part: The patched BIOS booted, the HDD auto detection reported zero size for my old 2,1GB HDD, but the startup screen reported the correct 2112MB.

Now to the bad part: The installed LINUX booted much slower than usual, the syslog reporting

- - - - - - - - - -

Feb 14 18:02:44 minieis kernel: hda: dma_timer_expiry: dma status == 0x21
Feb 14 18:02:44 minieis kernel: hda: error waiting for DMA
Feb 14 18:02:44 minieis kernel: hda: dma timeout retry: status=0x58 { DriveReady SeekComplete DataRequest }

- - - - - - - - - -

Sometimes the system even halted, the HDD LED lightened (it continued afterwards reporting the error above again, but I never had these problems with my old BIOS).

Well, I've done some research, tried all available BIOS and watched the bootup process and syslog closely: The error stated above occurs with BIOS 1.4, 1.5, 1.6b1 and 1.6b1 patched. So it's *no* problem of the patch itself!

BIOS 1.1 (which I had initially installed) and 1.2 are fine, but the syslog reports

- - - - - - - - - -

Feb 13 18:22:59 eis kernel: hda: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
Feb 13 18:22:59 eis kernel: hda: task_no_data_intr: error=0x04 { DriveStatusError }
Feb 13 18:22:59 eis kernel: hda: 4124736 sectors (2112 MB), CHS=1023/64/63, DMA

- - - - - - - - - -

even with these BIOS. With the others, too.

I'm getting the feeling that my old HDD is somewhat broken, though I'm confused about the slowed boot process and DMA error with BIOS 1.4 and above. Currently I'm testing the PC with BIOS 1.2 to be sure this one works.

Perhaps you can patch 1.1 and / or 1.2 which I have sent to you. To be honest: There's nothing of real importance to me in BIOS 1.4 and above. ;-)

I will try to get another HDD to test the 1.6b1 patched, though, and will report as soon as possible.

Oh, and before I forget: All BIOS from 1.6b1 and "above" report in the Flash utility, that they were not written for the computer at hand. I think this is because of the beta stadium of 1.6, but kind of irritating (but no problem at all).


Thank you again for your help. Please excuse me bad english.

René
edwin
The Hardware Archivist
Posts: 6286
Joined: Wed Mar 20, 2002 7:11 pm
Location: Netherlands
Contact:

sent the 1.1 and 1.2 patched as well. the bios string differs slightly that is why the flasher complains.
edwin/evasive

Do not assume anything

System error, strike any user to continue...
hanker
New visitors - please read the rules.
Posts: 12
Joined: Tue Feb 13, 2007 6:49 pm

Just wanted to tell that I just got this DMA error with the unpatched 1.2 BIOS. Seems the HDD *is* broken. Interesting that the error *always* occurs with BIOS 1.4+ and *sometimes* with 1.2 and perhaps 1.1.

I have just flashed the patched 1.1 and got no problems. Auto detection reported zero size, startup showed 2112MB, only the "usual" errors (without DMA) in the syslog. The computer will run for the night and I will look for DMA errors.

I will try the patched 1.6 as soon as I get another HDD.


René
cp
BIOS Guru
Posts: 1914
Joined: Mon Oct 21, 2002 9:07 pm
Location: Germany

just for your information: linux doesn't need to know and doesn't care about what the bios reports. It just get's to ask the HDD all by itself through the ide controller. so you just have to make sure the bios doesn't crash on detecting the HDD. you can do this by clipping the drive to 32GB with the appropriate jumper (refer to your hdd manual) and redetecting the drive (on autodetection answer 2 or Y). once the kernel is loaded you'll be able to use the full hdd size.
hanker
New visitors - please read the rules.
Posts: 12
Joined: Tue Feb 13, 2007 6:49 pm

Yes, I know. But as you mentioned above the problem with some boards is, that they do not even boot when a HDD bigger than 32GB is installed.

I have installed a 40GB HDD with a 32GB clip in another PC that had this problem, and LINUX did *not* detect it as 40GB, but 32GB. I thought this was normal.

When I remove the 32GB clip, the BIOS hangs at startup.

I am using a 2.4.X kernel.


René
edwin
The Hardware Archivist
Posts: 6286
Joined: Wed Mar 20, 2002 7:11 pm
Location: Netherlands
Contact:

It's only the newer kernels that can do this, but you have to tell the kernel specific NOT to look at the bios for the drive size.

If the board has an Award bios with this bug it's best to patch it anyway as we can also remove any DMA funnies too (where the bios sets the disk to an DMA mode the chipset does not support to begin with).

In the mean time we are suspecting that 2.1GB disk is half-dead, this:
http://www.seagate.com/ww/v/index.jsp?l ... 04090aRCRD
may confirm that...
edwin/evasive

Do not assume anything

System error, strike any user to continue...
cp
BIOS Guru
Posts: 1914
Joined: Mon Oct 21, 2002 9:07 pm
Location: Germany

Yes, I know. But as you mentioned above the problem with some boards is, that they do not even boot when a HDD bigger than 32GB is installed.
When I remove the 32GB clip, the BIOS hangs at startup.
yes. that's the so called award bug.
I have installed a 40GB HDD with a 32GB clip in another PC that had this problem, and LINUX did *not* detect it as 40GB, but 32GB. I thought this was normal.
no. that's not normal.
I am using a 2.4.X kernel.
kernels 2.4.18 and upward support LBA48. make sure you're using a recent kernel and all your problems will dissapear. the kernel even sets DMA modes if the bios does not. or you can force them with hdparm, well, after you checked they are available with hdparm first ofcourse :)
hanker
New visitors - please read the rules.
Posts: 12
Joined: Tue Feb 13, 2007 6:49 pm

cp wrote:
I have installed a 40GB HDD with a 32GB clip in another PC that had this problem, and LINUX did *not* detect it as 40GB, but 32GB. I thought this was normal.
no. that's not normal.
I am using a 2.4.X kernel.
kernels 2.4.18 and upward support LBA48. make sure you're using a recent kernel and all your problems will dissapear. the kernel even sets DMA modes if the bios does not. or you can force them with hdparm, well, after you checked they are available with hdparm first ofcourse :)
I have a 2.4.26-1. Funny things happen right here!

But that's not the problem at hand. I will have a look at it when I update *that* system (there is a patch for that BIOS already). Currently I don't need the extra 8GB, and: Never touch a running system. ;-)

I will test the HDD this afternoon.

René
hanker
New visitors - please read the rules.
Posts: 12
Joined: Tue Feb 13, 2007 6:49 pm

I just tested the HDD with BIOS 1.1 patched and 1.6B1 patched and the Seagate tool returned no errors.

Funny thing, that one...

Trying to get another HDD this weekend.

René
cp
BIOS Guru
Posts: 1914
Joined: Mon Oct 21, 2002 9:07 pm
Location: Germany

did you change any jumpers on the drive or the ide cabeling? never use the middle connector on a one drive setup: it's evil, it's forbidden and it leads to errors especially with DMA transfers.
hanker
New visitors - please read the rules.
Posts: 12
Joined: Tue Feb 13, 2007 6:49 pm

The drive (we are talking about the 2,1GB here, not my 32GB "problem") has three jumpers:

5-6 closed: master or standalone

all open: slave

3-4 (middle): cable select

3-4 and 5-6 closed: master with non ATA slave

It is installed as a single drive and thus 5-6 is closed.


The cable is a short standard IDE cable with only one connector, nothing special.

I just tried the CD-ROM cable (BIOS 1.6B1). It all seemed OK, but the

- - - - -

Feb 15 19:06:20 minieis kernel: hda: dma_timer_expiry: dma status == 0x21
Feb 15 19:06:35 minieis kernel: hda: error waiting for DMA
Feb 15 19:06:35 minieis kernel: hda: dma timeout retry: status=0x58 { DriveReady SeekComplete DataRequest }

- - - - -

happened shortly after bootup. Seems the problem is connected to the cable used, but the cable isn't the main problem.

But I found another interesting thing: When installing 1.6B1 LINUX activates ACPI, though my Mainboard does not support it. Is there a possibility that this is the problem? (I'm currently looking how to deactivate ACPI by hand, "acpi=off" and "pci=noapci" do not work at my eisfair system ( http://www.eisfair.org ).)


René, living with BIOS 1.1 and happy with it ;-)
hanker
New visitors - please read the rules.
Posts: 12
Joined: Tue Feb 13, 2007 6:49 pm

It's not the ACPI. Deactivated it, and got the old message. Got the DMA error with 1.1 for the first time, too.

Should get another HDD tomorrow.


René
Post Reply