Compiling ACPI v2.0 driver for Windows XP SP3 and Windows 2003 SP2 (x32/x64)

[Dave-H]

Thanks, I have to say that a lot of this stuff is way over my head!
So when the UAA driver first starts on boot it presumably isn't seeing the Realtek device.
However it does see it if it is then stopped and restarted, which seems very odd.

[Dietmar]

@Dave-H

I make a try with the more XP friendly Bios version 93cn27ww for the Flex 10 board.

Exact the same symptoms, no restart, sound only when UAA driver disabled and then enabled again.

And Sound lost again after next reboot.

I also make a try with a naked IDE XP SP3, which I install on another compi.

There, sound and restart work at once, but when put it into the Flex 10,

same again.

I think with @daniel_k , that this behavior is only because of its crazy DSDT

Dietmar

Edited May 14, 2022 by Dietmar

[Dave-H]

One other thing I've now found is that if I put the machine into standby, it appears to be impossible to wake it up again.
I had to resort to holding the power button down until it reset.
Also, and I've seen this several times now, if the disk check runs on boot, it appears to work OK, but then I get a 0x50 BSOD (PAGE_FAULT_IN_NONPAGED_AREA) apparently caused by dump_ntoskrn8.sys.
It then restarts OK.

[Mov AX, 0xDEAD]

On 5/13/2022 at 10:18 PM, daniel_k said:

Probably related to GPIO, again.

Related code in DSDT, HDA device:

Quote

Device (HDEF)
        {
            Name (_ADR, 0x001B0000)  // _ADR: Address
            OperationRegion (HDAR, PCI_Config, 0x4C, 0x10)
            Field (HDAR, WordAcc, NoLock, Preserve)
            {
                DCKA,   1, 
                Offset (0x01), 
                DCKM,   1, 
                    ,   6, 
                DCKS,   1, 
                Offset (0x08), 
                Offset (0x09), 
                PMEE,   1, 
                    ,   6, 
                PMES,   1
            }

 

@Dietmar

You can check PCI regs with RW Everything about PMEE and PMES bits when driver started and when not

Select HDA controller, offset 0x4C+ 0x9, bit 0 and bit 7

Intel PDF tells Power Management Control And Status at offset 0x54 for 110 and 370 chipsets, my 110 DSDT confirm it: 0x54 = 0x08, PMME=0 PMES =0

Quote

 Device (HDAS)
            Field (HDAR, ByteAcc, NoLock, Preserve)
            {
                Offset (0x54),
                Offset (0x55),
                PMEE,   1,
                    ,   6,
                PMES,   1
            }

Edited May 14, 2022 by Mov AX, 0xDEAD

[Dietmar]

@Mov AX, 0xDEAD

NoSound

Sound

[Dietmar]

@Mov AX, 0xDEAD

NoSound UAA no entries in Bus Relations

 

Sound UAA 2 entries in Bus Relations

HDAUDIO\FUNC_01&VEN_10EC&DEV_0269&SUBSYS_17AA3800&REV_1001\4&16ee679&0&0001
HDAUDIO\FUNC_01&VEN_8086&DEV_2882&SUBSYS_80860101&REV_1000\4&16ee679&0&0201

EnergyStatus always D0 (NoSound, Sound)

Dietmar

Edited May 14, 2022 by Dietmar

[Mov AX, 0xDEAD]

1 hour ago, Dietmar said:

@Mov AX, 0xDEAD

NoSound

 

0x54 is not recognized

[Dave-H]

Just to say that I've found a very crude workaround for the sound problem.
I've made a batch file with

devcon disable "PCI\VEN_8086&DEV_0F04*"
devcon enable "PCI\VEN_8086&DEV_0F04*"

which runs from my startup folder.
This works, and I now have sound without actually having to do anything manually.
Not an ideal permanent solution though!

[Mov AX, 0xDEAD]

my mistake, 0x4c+0x9 = 0x55, field definition is right on Flex DSDT

[Dietmar]

@Mov AX, 0xDEAD

I think, that it is not allowed,

that in UAA Bus Relations are 2 Entries.

Only one is allowed(?)

Dietmar

[Dietmar]

@Dave-H

This can also be a working solution for @Damnation and his USB problem.

Can you write short Tutorial about to get this batch file working under XP

Dietmar

[Dave-H]

Well it doesn't really need a tutorial, at least not for anyone here I would have thought.
As long as you have the Devcon program, which is readily downloadable, you just need to put it somewhere where it can be found, such as the \Windows\System32 folder, and put those commands into a text file and save them as a .bat batch file.
@Damnation would obviously have to put in the appropriate ID for his problem device.
How you run it doesn't really matter I wouldn't have thought, you could run it manually after boot if you want to, or from the registry or just put it in the startup folder as I did.
I'm not sure whether the most recent versions of Devcon work on XP, so just in case an XP compatible version is now hard to find, I've attached mine.

DEVCON.zip

[Dietmar]

@Mov AX, 0xDEAD

A look in the old XP thread shows me, that the Baytrail board Q1900m has exact the same problem with the UAA driver. So maybe it is a hardware problem of Intel Baytrail chips

Dietmar

Edited May 15, 2022 by Dietmar

[Dietmar]

@Mov AX, 0xDEAD

This "small" bug with sound may be a really big bug for all the Baytrail chips.

Because the Baytrail chips have one and the same the Realtek sound, USB, cpu all together in one chip,

called SoC.

In Linux they solved the problem with

intel_idle.max_cstate=1

https://bugzilla.kernel.org/show_bug.cgi?id=109051

Dietmar

PS: The Linux Bug report mentioned also the power shutdown problem on all Baytrail Soc, brrr..

https://bugzilla.kernel.org/show_bug.cgi?id=70931

 

Edited May 15, 2022 by Dietmar

[Mov AX, 0xDEAD]

@Dietmar

On you screen PMEE=1 (0x55=0x1) with sound and without

PMEE=0 after loading windows to desktop with enabled or disabled UAA Controller on my H110

PMEE=1 can be only after seting this bit manualy, PMEE automaticaly cleared when i reenable UAA device in device manager

Also on your "NoSound" screen RW Everything still detects CodecID=10EC0269, it means HDA bus is active, but maybe RW uses different way to look for HDA devices. Reselect PCI device in RWE to re-read actual state after disable/enable in Device Manager

When UAA controller disabled in device manager, RW Everything stop detecting HDA devices

Edited May 15, 2022 by Mov AX, 0xDEAD