Acer M3A78-EM motherboard

[dgbisse]

I have a new Acer M3A78-EM motherboard with a triple core AMD Phenom processor. After much messing around, I got it to work in ACPI SMP mode, using /SMP /PIC /FS /CD with ACPI 3.11 updated by ACPI12112008.

If I use no parameters, it works fine, in single processor mode (SMP 1.04a kernel). It also works perfectly, using the OS2APIC.PSD, and the old APM drivers, although it won't suspend (I haven't tried suspend with ACPI, yet).

If I try /APIC, it just hangs loading the USB drivers. If I disable the USB drivers, it goes further, but eventually takes a TRAP000d in the kernel (it appears that there were many traps, before that came up on the screen). I will attach the logs, as collected by the ACPI logs collector.

[dgbisse]

When I tried to upload the logs file, the machine took a trap0008. I tried to get a photo of it, but they didn't turn out too good. Attached anyway.

[pasha]

what is what using psd=acpi.psd /SMP /APIC ? And downgrade pls vpic.sys

[eco]

e-co:

AMD chipset:

​http://www.asus.com/products.aspx?modelmenu=2&model=2252&l1=3&l2=149&l3=639&l4=0

it's strange that you can boot the system with /SMP /PIC

[dgbisse]

Updated to 3.13, using acpifu-20081127.wpi, and will upload new logs.

VPIC.SYS is, currently, dated Nov. 12 2008.

[dgbisse]

I updated to ACPI 3.14, using the WPI file. New log file uploaded. There was a PROCDUMP created, when I was creating the log files. That too is uploaded (SMP processor mode), along with POPUP:OG.OS2.

Now, it runs, using /SMP /PIC /CD.

[dgbisse]

Could not upload PROCDUMP - too big...

I will hold it. If you want it, please tell me how to get it to you.

[dgbisse]

Files attached, as requested in the news groups: noparms_acpi-log-Model-of-your-PC.ziq - No parms on ACPI.PSD. smp_pic_acpi-log-Model-of-your-PC.ziq - /SMP /PIC on ACPI.PSD smp_APIC_hang.jpg - /SMP /APIC on ACPI.PSD, with DEBUG ACPI.PSD

Hope this helps...

[pasha]

What is options are use with /APIC ?

[eco]

e-co:

homepage of MB: ​http://www.asus.com/products.aspx?modelmenu=2&model=2252&l1=3&l2=149&l3=639&l4=0

SYM53C8XX - do you need this scsi adapter?

if you need SCSI then use PIC SMP: ​http://ecomstation.ru/projects/acpitools/?action=acpi-scsi

[dgbisse]

Replying to pasha:

What is options are use with /APIC ?

Just /SMP /APIC, or /SMP /APIC /CD, or /SMP /APIC /CD /FS.

All have the same results, a hang at the boot logo. If I do Alt-F2, the hang is when USBOHCD.SYS tries to load (OHCD loads before EHCD).

[dgbisse]

Replying to eco:

e-co:

homepage of MB: ​http://www.asus.com/products.aspx?modelmenu=2&model=2252&l1=3&l2=149&l3=639&l4=0

SYM53C8XX - do you need this scsi adapter?

if you need SCSI then use PIC SMP: ​http://ecomstation.ru/projects/acpitools/?action=acpi-scsi

I have tried most of what is posted on those pages. The hang in ACPI has nothing to do with the SCSI card (which I need to run my scanner). I can REM the device driver, and remove the card, and I still get the same hang, at USBOHCD.SYS, when I try to use /APIC. /PIC works fine, but the problem should be fixed, eventually.

[eco]

e-co:

OK, you have SCSI then SMP APIC doesn't work for you (for drivers of SCSI controller)

So.. stay with /PIC /SMP

Confirm that we can close the ticket

[pasha]

Please download experimental ACPI build from Mensys site:

• Experimental build for you:

ACPI-APIC-BAT.ZIP

• Read instructions: ​http://ecomstation.ru/projects/acpitools/?action=testcase

[eco]

e-co:

USB Legacy = Off in BIOS Setup, if this doesn't help then disable USB controllers drivers, try boot again.

[dgbisse]

Replying to eco:

e-co:

OK, you have SCSI then SMP APIC doesn't work for you (for drivers of SCSI controller)

So.. stay with /PIC /SMP

Confirm that we can close the ticket

I have SCSI, but that has NOTHING to do with the problem. I can remove the SCSI card, and I still have the same problem. NO, you may NOT close this incident, until it is fixed.

[dgbisse]

Replying to pasha:

Please download experimental ACPI build from Mensys site:

• Experimental build for you:

ACPI-APIC-BAT.ZIP

• Read instructions: ​http://ecomstation.ru/projects/acpitools/?action=testcase

Tried that. Picture smpAPIC!nodSAP0x10.jpg is with /SMP /ACPI /!NOD /SAP:0x10 as suggested, somewhere. SMPapicCD.jpg is with parameters /SMP /APIC. It still works with /SMP /PIC.

[dgbisse]

see text

[dgbisse]

Replying to eco:

e-co:

USB Legacy = Off in BIOS Setup, if this doesn't help then disable USB controllers drivers, try boot again.

I have tried it both ways, as well as with all of the other ACPI options in BIOS. Leaving USB Legacy on, causes other problems with my external USB hard drive.

[pasha]

Attach log with /SMP /PIC. copy acpica$ acpi.log after boot.

[dgbisse]

New result, while using the contents of: ACPI-APIC-BAT.ZIP.

With this code, I get a Trap0008, when I try to start Virtual PC 5.1, with a win2K guest, but it fails so early, that I am sure that it doesn't matter what the guest is. I am not able to report the rest, because the Trap message is there, only long enough to make out that it says Trap0008, then the screen is written with some meaningless garbage. Going back to the "official" 3.14 release, allows me to run VPC again.

[dgbisse]

Logs with ACPI-APIC-BAT.ZIP contents loaded

[dgbisse]

POPUPLOG.OS2, produced by esysinfo

[dgbisse]

Replying to pasha:

Attach log with /SMP /PIC. copy acpica$ acpi.log after boot.

Logs attached, as requested. I also added POPUPLOG.OS2 that esysinfo produced. There is also a PROCDUMP, but it is too big to upload. This is with the contents of ACPI-APIC-BAT.ZIP loaded.

[pasha]

Need normal acpi.log from ACPI-APIC-BAT.ZIP

[pasha]

Please download experimental ACPI build from Mensys site:

• Experimental build for you:

ACPI315PRE.ZIP

• Read instructions: ​http://ecomstation.ru/projects/acpitools/?action=testcase

Need log from /SMP /PIC and photo of screen for /SMP /APIC

[dgbisse]

Replying to pasha:

Please download experimental ACPI build from Mensys site:

• Experimental build for you:

ACPI315PRE.ZIP

• Read instructions: ​http://ecomstation.ru/projects/acpitools/?action=testcase

Need log from /SMP /PIC and photo of screen for /SMP /APIC

/SMP /APIC still just stops at loading USBOHCD.SYS.

/SMP /APIC /CD does the same.

/SMP /APIC /CD /FS causes a repeating trap, which eventually stops with that is in Test2.jpg.

/SMP /ACPI /!NOD /SAP:0x10 (as suggested earlier) gets further, but eventually stops with Test1.jpg.

/SMP /PIC still works.

Log collection with /SMP /PIC will follow.

[dgbisse]

I have made a change to this machine. The triple core processor is out, and a quad core is in.

[dgbisse]

I have tried ACPI v3.16 (acpi316-07-23.zip) with this machine. It took a while to realize that it is ACPI that is causing system hangs, mostly when I attempt to display pages in Firefox 3.5.2, but at other times too (including when I am not even using the machine). There are no logs, and no indication of a problem, other than nothing works. The mouse won't move, the desktop is frozen, the LAN will not respond. The only things that work, are the reset button, or the power off button (requires the 5 second power off). None of this stuff happens when I don't use ACPI. I did see similar problems with earlier versions of ACPI, but they were not as consistent, and I never related them to ACPI. Now, I have no doubt, that ACPI is causing the problem. With ACPI, and FF 3.52, I can cause the hang in about 20 minutes. Previously, it would take about 4 hours. Without ACPI, I can run for more than 7 days, before anything odd happens.

[pasha]

Try add /TMR and/or read FAQ about set time oprions for FF

[dgbisse]

Good news. I tried version 3.17, from ACPI317-2008-08-24.ZIP. I am now almost 24 hours into a boot, with no trouble. This is better than I was able to do with 3.14, 3.16, or the initial release of 3.17.

Thanks...

[dgbisse]

Hmmm. Something has changed. In my BIOS settings, I have a choice of: ACPI 2.0 Support enabled/disabled The description is: Add additional tables as per ACPI 2.0 specifications. Also: ACPI APIC support enabled/disabled The description is: Include ACPI APIC table pointer to RDST pointer list. If I disable both, the system boots, but initializes only ONE (of 4) processor. ACPI 3.14 initializes 4 processors, as does OS2APIC.PSD. If I enable only ACPI 2.0 Support, only one processor is initialized, but I get a Trap 000e at about the time when the desktop should appear. If I enable only ACPI APIC support, 4 processors are enabled, but I get a repeating Trap 000d, alternating with Trap 0008, at about the time when the desktop should appear. ACPI 3.14 works, for a while. If both are enabled, 4 processors are enabled, I get the repeating Trap 00d/Trap008. ACPI 3.14 works, for a while.

The above report, about 3.17 working better, was based on it running only one processor, with both settings off. I didn't notice that it was only one processor, at the time.

Pictures of the trap screens will be loaded later.

[dgbisse]

Tried 3.18S. This is definitely better than anything since 3.14, but I have only run it for about 15 minutes, so far. More later...

[dgbisse]

I updated to ACPI 3.18, as distributed in the file ACPI318FULL.ZIP. This works better than anything I have tried, since v3.14. Testing will continue...

[dgbisse]

I should have noted that this machine still needs to run with /SMP /PIC. It will not boot when /APIC is used.

[pasha]

Please download experimental ACPI build from Mensys site:

• Experimental build for you:

acpi319ForDesktop.zip

• Read instructions: ​http://ecomstation.ru/projects/acpitools/?action=testcase

[dgbisse]

Sorry, but I completely missed your previous update.

Updated to acpi319ForDesktop.zip

A quick test with /SMP /APIC or /SMP /APIC /CD causes a trap. Photo uploaded 2010-12-29_21-52-22.jpg

With /SMP /PIC it seems to work as with v 3.18.

[pasha]

Can you attach acpi log from 3.19 with /SMP /PIC ? copy acpica$ acpi.log

[dgbisse]

acpi.log uploaded, as requested, using the contents of ACPI319CR4.ZIP.

Is there any way that I can be notified that this incident is updated? I had no clue that a question was asked, until I happened to be looking here for another problem.

[pasha]

Please download experimental ACPI build from Mensys site:

• Experimental build for you:

acpi319PollingFullOnly.zip

• Read instructions: ​http://ecomstation.ru/projects/acpitools/?action=testcase

[dgbisse]

I tried acpi319PollingFullOnly.zip with no change. It runs okay with /SMP /PIC, but hangs at USBOHCD.SYS when I try /SMP /ACPI /CD.

[pasha]

if you are remming USBOHCD.SYS, is boot right? And, do you have physical floppy drive?

[dgbisse]

Yes, there is a floppy drive.

I also changed Polling Part to Polling Full in M:\OS2\BOOT\acpi.cfg.

I tried REMing all of the USB driver statements. It did get further, but crashed while loading UNIAUD32.SYS, or UNIAUD16.SYS (I couldn't tell by watching it), so I REMed those two entries as well. See ACPI1.JPG.

Then, it stopped at a later point. See ACPI2.JPG.

I then took all of the REMs out, and changed the command to "/SMP /PIC" and it seems to work.

[pasha]

is uniaud last from netlabs?

[pasha]

feedback pending -> mensys stage

[dgbisse]

I have since replaced that install with eCS 2.1, but UniAud? was either the latest, or the one previous.

[pasha]

Please download experimental ACPI build from Mensys site:

• Experimental build for you:

acpi319for105-2011-05-30.zip

• Read instructions: ​http://ecomstation.ru/projects/acpitools/?action=testcase

[dgbisse]

This is with a new install of eCS 2.1.

For acpi319for105-2011-05-30.zip:

Using "PSD=ACPI.PSD /SMP /PIC /VBE", it works.

Using "PSD=ACPI.PSD /SMP /APIC /VBE", it traps as shown in 2011-06-01_10-26-34.JPG.

[pasha]

Please download experimental ACPI build from Mensys site:

• Experimental build for you:

acpi319APICFix-20110605.zip

• Read instructions: ​http://ecomstation.ru/projects/acpitools/?action=testcase

[dgbisse]

acpi319APICFix-20110605.zip

Works with PSD=ACPI.PSD /SMP /PIC /VBE

Traps with PSD=ACPI.PSD /SMP /APIC /VBE. See M3A78-EMTrap.jpg.

[pasha]

Can you attach log frov PSD=ACPI.PSD /SMP /PIC /VBE ?

[dgbisse]

acpidaemon.log attached.

[dgbisse]

I just tried ACPI3_19_14.exe. This machine is now running properly, with no parameters on the PSD line. That has never happened before.

Well done.

[dgbisse]

I should have noted that APIC is active, and working.

[David Azarewicz]

Thanks for testing.