Solved (was: snd-hda-intel works only with acpi=off)

[ijontichy]

Hi,

on my Toshiba Satellite P100-119, lspci gives me the following info about my soundcard:

Code: Select all

00:1b.0 Audio device: Intel Corporation 82801G (ICH7 Family) High Definition Audio Controller (rev 02)

My /etc/modules.d/alsa is

Code: Select all

alias /dev/mixer snd-mixer-oss
alias /dev/dsp snd-pcm-oss
alias /dev/midi snd-seq-oss

alias snd-card-0 snd-hda-intel
alias sound-slot-0 snd-card-0

This works fine if when booting with acpi=off.
However, when booting with acpi enabled, the soundcard is silent.

dmesg gives me the following:

booting with acpi=off:

Code: Select all

PCI->APIC IRQ transform: 0000:00:1b.0[A] -> IRQ 22
PCI: Setting latency timer of device 0000:00:1b.0 to 64

booting with acpi enabled:

Code: Select all

ACPI: PCI Interrupt 0000:00:1b.0[A] -> GSI 22 (level, low) -> IRQ 21
PCI: Setting latency timer of device 0000:00:1b.0 to 64

So i suspected that acpi might cause an IRQ routing problem, but
using pci=noacpi at boot gives no difference (same dmesg, no sound).

In order to get acpi working, i had to fix my dsdt. Apart from the sound
problem, acpi works fine now.
Also, the sound modules load without errors, alsamixer works etc., just i get not sound.
The sound related part of the dsdt seems to be this:

Code: Select all

        Device (AMW0)
        {
            // Name (_HID, "*PNP0C14")  
            Name (_HID, EisaId("PNP0C14"))
            Name (_UID, 0x00)
            // Name (_WDG, Buffer (0x3C)
            Name (WDG, Buffer (0x3C)
            {
                0xA7, 0x1D, 0x85, 0x2E, 0x53, 0xD0, 0x5F, 0x49, 
                0x9D, 0xFA, 0x1A, 0x4A, 0xD6, 0x2E, 0x6A, 0x86, 
               ...

Mybe i introduced an error here?

Any help would be highly appreciated.

Christian

[rlittle]

There is someone else!

I have a Toshiba Satellite P100 (JR500E) also, and have exactly the same problem. acpi=off is the only thing that's worked so far, so thank you 100 times for that! At least I know that I haven't set up ALSA wrong 37 times in a row...

I have exactly the same messages that you posted. Sorry to not have any more insight into the prob, but if I do I'll post it here.

Question: did you fix your dsdt "by hand" or did you find an already fixed copy? (I fixed the dsdt on a laptop and couple of years ago and have thankfully forgot everything I learned)

[ijontichy]

Hi,

i had to fix my dsdt by hand (without too much understanding).
I seems to work satisfactorily; i can read battery state, temperature
and get button events (using the vanilla kernel 2.6.16.1):
This is what i changed:

Code: Select all

241a242
>     External (CFGD)  // Hack without understanding
600a602
>           External(\PDC0) // Hack without understanding
610a613
>           External(\PDC1) // Hack without understanding
5702,5703c5705,5706
<                 Z001, 
<                 Z001, 
---
>               // Z001, Hack without understanding
>                 // Z001, dito  
6011,6012c6014,6019
<             }
<         }
---
>               Else
>               {
>                  Return(Zero)
>               } 
>           }
>       }
6016c6023,6024
<             Name (_HID, "*PNP0C14")
---
>         // Name (_HID, "*PNP0C14")
>           Name (_HID, "PNP0C14") // Hack withour real understanding
6018c6026,6027
<             Name (_WDG, Buffer (0x3C)
---
>             Name (_WDG, Buffer (0x3C) // Hack withour real understanding
>                 //                     Name (WDG, Buffer (0x3C)

Sound is still not working with acpi.

Also, i cannot get a usable X configuration under Gentoo.
The nvidia driver gives oopses (tried several versions) and the nv driver
says that does not detect any devices.
I tried SuSE 10.1 beta8 for comparison, where i get at least X working
with nv.

Was your P100 also equipped with an ``express media player''?
Interestingly enough this was running under Linux! So i knew it
would be possible to get sound under Linux.
On my laptop, this was installed on a hidden partition.
fdisk showed me two partitions sda1 and sda4 (no sda2, sda3),
gpart showed me the two hidden partitions sda2, sda3.
Somewhere on these, the expressed media player resided, but i was
not able to mount them (i hoped that i would be able to copy some
parts of the config).
Next i looked at the recovery CD of the express media player.
It is based on syslinux (which i dont understand).
Normally, the recovery cd boots into a menu with few options
(reinstall, delete all partitions, or quit).
But if one boots this recovery cd and plays with the keyboard during boot,
one gets into the grub prompt an can boot into the shell and look around:-)
Maybe this way one can extract some some usefull information.

Looking at the changelogs from kernel.org, i get the impressions that
currently there is some progress with the snd-hda-intel module.
So i would be satisfied, if i could get X working under Gentoo (nv would
suffice).

Christian

[rlittle]

I'll post an update when I'm on my laptop, but I had no real trouble with X (however, I'm not sure why, because a friend had the same problem). I installed via stage 3 and the Live CD, but I had to boot with the option

Code: Select all

gentoo-nofb -nox

because the Live CD couldn't recognize the video card properly and it would freeze while starting Gnome. After I finished the install from the command line and had everything booting and networked (the e1000 card anyway, wireless was a different story ) I installed X and Gnome and got them up and running (well.... let's say "not crashing", not "running"), then I emerged the latest and greatest nvidia driver (nvidia-kernel, nvidia-glx and nvidia-settings) and (after struggling with xorg.conf for 8 hours) it worked the first time. I'll post my xorg.conf as soon as I can. Right now I'm still having trouble getting the wheel mouse working on my extra USB mouse, but I think I have less-than-perfect settings for "CorePointer".

Thanks a ton for your dsdt fixes! If your can detect bettery levels, then you're my hero!

One odd thing, my friend also installed Fedora Core on the laptop (it's identical to mine, we got them from work) and he had sound with no problem! (however, he couldn't get X working at all, and he couldn't get the e1000 NIC working either. )

Maybe between the two of us, we can get a working system!

Edit - almost forgot, if you're going for the nvidia-kernel driver, remove all references to nv and DRI from your kernel.

[ijontichy]

Hi,

your xorg.conf would be indeed very usefull for me (including the exact version
of the nvidia driver and your kernel).
Unfortunately out laptops are not identical.
So, probably you cannot apply my dsdt fixes verbatim.

Also, possibly the lack of sound comes from the fact that
my dsdt fixes are not yet perfect.


Christian

[rlittle]

Here you go:

Nvidia versions are:

nvidia-kernel-1.0.8178-r1
nvidia-glx-1.0.8178-r1
nvidia-settings-1.0.20051122-r3

and my kernel is 2.6.15-r7

# uname -a
Linux <hostname> 2.6.15-gentoo-r7 #15 SMP PREEMPT Thu Mar 30 21:02:32 Local time zone must be set- i686 Genuine Intel(R) CPU T2400 @ 1.83GHz GenuineIntel GNU/Linux

(apparently I need to set my timezone correctly. )


And here is my xorg.conf. There are a couple of things to be aware of. I setup the touchpad by installing the synaptics touchpad driver ( emerge x11-drivers/synaptics ). I also have a "Modeline" in there which I'm sure is correct, but I'm not using properly. I think I need to add a line refering to "1440x900" into the "Screen" section somewhere. However, it should cause no problems for anyone. Also, notice that the "dri" module had been commented out.

Code: Select all

Section "ServerLayout"
        Identifier     "X.org Configured"
        Screen         "Screen0"
        InputDevice    "TouchPad0" "AlwaysCore"
        InputDevice    "Mouse0" "CorePointer"
        InputDevice    "Keyboard0" "CoreKeyboard"
EndSection

Section "Files"
        RgbPath      "/usr/lib/X11/rgb"
        ModulePath   "/usr/lib/modules"
        FontPath     "/usr/share/fonts/misc/"
        FontPath     "/usr/share/fonts/TTF/"
        FontPath     "/usr/share/fonts/Type1/"
        FontPath     "/usr/share/fonts/CID/"
        FontPath     "/usr/share/fonts/75dpi/"
        FontPath     "/usr/share/fonts/100dpi/"
EndSection

Section "Module"
        Load  "glx"
        Load  "record"
        Load  "extmod"
        Load  "dbe"
#       Load  "dri" #don't want this with the nvidia binary driver
        Load  "xtrap"
        Load  "freetype"
        Load  "type1"
EndSection

Section "InputDevice"
        Identifier  "Keyboard0"
        Driver      "kbd"
EndSection

Section "InputDevice"
        Identifier  "TouchPad0"
        Driver      "mouse"
        Option      "Protocol" "auto"
#       Option      "Protocol"  "auto-dev"
        Option      "Device" "/dev/input/mice"
        Option      "LeftEdge"      "1700"
        Option      "RightEdge"     "5300"
        Option      "TopEdge"       "1700"
        Option      "BottomEdge"    "4200"
        Option      "FingerLow" "25"
        Option      "FingerHigh"        "30"
        Option      "MaxTapTime"        "180"
        Option      "MaxTapMove"        "220"
        Option      "VertScrollDelta" "100"
        Option      "MinSpeed"  "0.09"
        Option      "MaxSpeed"  "0.18"
        Option      "AccelFactor"       "0.0015"
        Option      "SHMConfig" "on"
EndSection


Section "InputDevice"
        Identifier  "Mouse0"
        Driver      "mouse"
        Option      "Protocol" "IMPS/2"
        Option      "Device" "/dev/input/mice"
        Option      "Buttons" "3"
        Option      "ZAxisMappping" "4 5"
EndSection


Section "Monitor"
        Identifier   "Monitor0"
        VendorName   "Monitor Vendor"
        ModelName    "Monitor Model"
        VertRefresh   60
        Modeline     "1440x900" 96  1440 1504 1536 1760  900 903 906 912  -HSync -VSync
EndSection

Section "Device"
        Identifier  "Card0"
        Driver      "nvidia"
        VendorName  "nVidia Corporation"
        BoardName   "Unknown Board"
        BusID       "PCI:1:0:0"
EndSection

Section "Screen"
        Identifier "Screen0"
        Device     "Card0"
        Monitor    "Monitor0"
        DefaultDepth       24
        SubSection "Display"
                Viewport   0 0
                Depth     1
        EndSubSection
        SubSection "Display"
                Viewport   0 0
                Depth     4
        EndSubSection
        SubSection "Display"
                Viewport   0 0
                Depth     8
        EndSubSection
        SubSection "Display"
                Viewport   0 0
                Depth     15
        EndSubSection
        SubSection "Display"
                Viewport   0 0
                Depth     16
        EndSubSection
        SubSection "Display"
                Viewport   0 0
                Depth     24
        EndSubSection
EndSection

Thanks for pointing out that our machine may differ. Do you have a Phoenix BIOS or Toshiba?

[vPat]

Iam also having this trouble with kernel 2.6.16-r1.

I have found this issue here, when i try compling sound into the kernel i still get no sound. Iam now going to compile them as modules and try with acpi=off.

.Pat

[ijontichy]

Hi,

concerning graphics, perhaps we have different graphics chips.
rlittles xorg.conf does not work for me. The system freezes if i try
to start X with nvidia.
Also from the nvidia linux forum it got the information, that nvidia has announced
to integrate suppoert for this chip into the next release.
My laptop has a NVIDIA GeForce Go 7600, which also seems to be currently not
supported by xorg 6.8.

But, in principle it is possible to get X it working with nv. I had no X problems
with SuSE 10.1 beta8; i guess they use xorg 7.0.

Concerning sound, in my case the problem also occurs with kernels prior to 2.6.16
(i tried 2.6.14.2 and 2.6.15.1), so it has nothing to do with the issue mentioned by
vPat.
The error seems to occur during the pci device scan, before loading any sound module.
I noticed, that the contents of /sys/bus/pci/devices/000000\:00\:1b.0
depends on whether acpi is enabled or not.
Somehow acpi is involved, even if i boot with pci=noacpi.
In my eyes, this is a kernel bug.

Christian

P.S. The bios on the P100-119 is from Phoenix.

[ijontichy]

Hi,

meanwhile i could solve the problem by applying simple
fixes to the dsdt. (I did some Bios upgrades in the meantime,
so depending on your bios version (and iasl version) the fixes might
be slightly different).

1. Get rid of iasl error and warnings:
- Replace all occurences of _T_0 by some arbitrary variable name
(_T_0 is a reserved word for iasl).
- Complete the return path for method BTST (just ad an additional
``Else { Return(Zero) }'')
- Replace ``Name (_HID, "*PNP0C14")'' by `` Name (_HID, "PNP0C14")''.

2. Get sound (+acpi) working:

Code: Select all

    Scope (\_SB)
    {
        Method (_INI, 0, NotSerialized)
        {
            If (DTSE)
            {
                TRAP (0x47)
            }

            Store (0x07D0, OSYS)
//             If (CondRefOf (_OSI, Local0))
//             {
//                 If (_OSI ("Linux"))
//                 {
//                     Store (0x03E8, OSYS)
//                 }
//                 Else
//                 {
//                     Store (0x07D1, OSYS)
//                     If (_OSI ("Windows 2001 SP2"))
//                     {
//                         Store (0x07D2, OSYS)
//                     }

//                     If (_OSI ("Windows 2001.1"))
//                     {
//                         Store (0x07D3, OSYS)
//                     }

//                     If (_OSI ("Windows 2001.1 SP1"))
//                     {
//                         Store (0x07D4, OSYS)
//                     }

//                     If (_OSI ("Windows 2006"))
//                     {
//                         Store (0x07D6, OSYS)
//                     }

//                     If (LAnd (MPEN, LEqual (OSYS, 0x07D1)))
//                     {
//                         TRAP (0x3D)
//                     }
//                 }
//             }
	      Store (0x07D6, OSYS)  // Fake Windows 2006
        }

Now i have both sound and acpi.
Just the sound volume is a bit low (but it was also low the same with apci=off).


Christian

[ijontichy]

Using non-kernel alsa-driver-1.driver-1.0.14_rc1 gives full sound volume.

So finally i have both: ACPI + sound fully working

[mentorek]

Hello,

It does work for me too Thanks for that solution. Have you emailed the corrected DSDT to Toshiba? Maybe they'll correct the DSDT entries in the next BIOS update?
That would made our life easier

EDIT: The PCM Volume control is now working (after rebooting machine).

Cheers,
Mik

[boroshan]

I can't make this work, and I don't understand why it doesn't.

I've compiled the new DSDT into the kernel using the custom dstd file options:

Code: Select all

CONFIG_ACPI_CUSTOM_DSDT=y
CONFIG_ACPI_CUSTOM_DSDT_FILE="/home/nick/toshiba/dsdt.hex"

I used this approach because it worked well with my previous lappy. The settings show up in /proc/config.gz, so I know the modified kernel is running. I've fixed the dsl file as per this thread. And yet nothing happens.

Which is weird, because sound still works with acpi=off, and (if I understand the hack properly) all we're doing is removing the part of the ACPI definition that affects sound - so it should have the same effect.

Has anyone got this working with a PSPA3E machine?

[rokoS]

I will try on my PSPA3E tonight.
I have one question. Does your Nvidia 7600 GPU fan work under linux. Because mine works only when i boot with acpi=off, and only with full speed.

Rok

[boroshan]

I will try on my PSPA3E tonight.
I have one question. Does your Nvidia 7600 GPU fan work under linux. Because mine works only when i boot with acpi=off, and only with full speed.

It seems to. I'm embarrassed to admit it, but I'm not sure which fan is which. That said, I've noticed both fans operating, so I think they both work.

[rokoS]

GPU fan is one on the left side. CPU is one on the back of the machine.

[boroshan]

I've certainly seen it working in the past. Can't say when last however, so recent kernel changes may have broken it.

[rokoS]

ON toshiba P100-115 (PSPA3E), sound works now. And you know what the GPU fan started to work also . No overheating anymore for my computer . It works all the time but is better then before when it didnt work. But now i think that CPU fan is not working properly.
I had to comment a lot of things in DSDT, besides the one that were mentioned in this thread.

boroshan, could you send me your DSDT so that i could see what were your changes and if it differs a lot from my toshiba.

[boroshan]

here you are. I've not done much to it beyond the basics

... and thanks

[thorheimdall]

I have a Toshiba P100-MA2 and I also have the same problems described here in this thread. With acpi=on, there is no sound from the sound card (hda-intel, conexant), and the nvidia 7600 fan never turns on (CPU fan starts and stops according to the CPU temp). With acpi=off, the GPU and CPU fans are always turned on, and the sound card works properly.

How should I proceed to compile the DSDT in the kernel? Is there some things I must/should change in the DSDT according to my laptop model?

Any help would be appreciated!

[boroshan]

How should I proceed to compile the DSDT in the kernel? Is there some things I must/should change in the DSDT according to my laptop model?

Probably the best place to start is the Gentoo wiki page. You might also find the sourceforge ACPI For Linux page useful.

If you get stuck, report back and I'll see what I can do

[thorheimdall]

Probably the best place to start is the Gentoo wiki page. You might also find the sourceforge ACPI For Linux page useful.

If you get stuck, report back and I'll see what I can do

Thank you very much boroshan! I'll take a look at those pages!

[thorheimdall]

I fixed the DSDT and now the sound card works with acpi=on, and the GPU fan is turned on all the time at a low speed. The DSDT fixes are basically the same as the ones suggested by ijontichy (replacing all occurrences of _T_0, "*PNP0C14"->"PNP0C14", and disable OS verification).

Thank you again for your help boroshan!

[Kenji Miyamoto]

Would this be the same sound chipset used in the Satellite A100? I have to max out the volume to get just enough sound out of those ear canal phones. Any lshw or lspci output regarding this problem would be helpful!

[boroshan]

Thank you again for your help boroshan!

Thanks for yours! I've obviously messed up something basic. Time to go back to basics and see what's wrong.

Would this be the same sound chipset used in the Satellite A100? I have to max out the volume to get just enough sound out of those ear canal phone. Any lshw or lspci output regarding this problem would be helpful!

Try this:

Code: Select all

cat /proc/asound/card0/codec#0 | head

I get this:

Code: Select all

Codec: Conexant CXT5047
Address: 0
Vendor Id: 0x14f15047
Subsystem Id: 0x1179ff31
Revision Id: 0x100000
Default PCM:
    rates [0x40]: 48000
    bits [0x2]: 16
    formats [0x1]: PCM
Default Amp-In caps: N/A

@rokoS: I just noticed I posted the hex version of the DSDT table. I'll post the dsl as soon as Zen's ftp server comes back up.

Edit:
done that

[Pinaraf]

Hi

I just tried this solution on the laptop of a friend, who is having this problem too (Toshiba Satellite P100-197, as far as I remember).
I updated the BIOS, no problem here.
I then changed the DSDT, but the kernel stopped booting :/
I restored the original DSDT, and did only the changes needed for the compiler to accept the DSDT : change "*PNP0C14" into "PNP0C14".... and it didn't boot either !

Do you have any suggestion ?

Thanks
Pierre