Release for Toshiba Satellite P10x users: modded BIOS to try and activate Conexant sound chip, already working on three moddels Options

[geiman]

Oh sorry, didn't realized you needed that too. Every other os x dvd I've tried installs the boot loader correctly, and it also recognizes windows and adds it to the boot loader. However, if you still want to use that same disc, I've had luck with either osl2000 boot manager, or Grub. Grub is probably the easiest since its free, but if you can find a copy of osl2000, it works good too. I'm pretty sure you can install the OS X boot loader also, and then add windows manually, but I can't think of how you do it off the top of my head.

[maranbrazil]

geiman, i will download the leo4all v3 iso.

thks for help

[geiman]

the iAtkos discs are also pretty good, in my opinion. I am using iAtkos r3, and after install, I do not need to install anything extra to get any part of my laptop working, minus wifi and of course sound. And it seems to be the fastest one I've tried yet, and I have tried many.

[geiman]

I just have a quick question for bugs, or whoever has any insight as to the exact problem with our bios's; Why exactly is it that windows can activate whatever is needed to have sound, but Linux/OS X cannot?

[mentorek]

I just have a quick question for bugs, or whoever has any insight as to the exact problem with our bios's; Why exactly is it that windows can activate whatever is needed to have sound, but Linux/OS X cannot?

Toshiba Satellite P100 series has a buggy DSDT in BIOS. This table holds all the information about the hardware, how and when to initialize it. What information has to be provided for certain OS's and how it should be provided.
In Linux there is a patch for the kernel that allows you to load corrected DSDT when starting up the system or "say" to BIOS that the system is not Linux (or something like that). The latter is widely used now.

[BugsB]

[..] or "say" to BIOS that the system is not Linux (or something like that). The latter is widely used now.

in grub.conf, you would just add this parameter to the kernel line, like this:

CODE

acpi_os_name="Microsoft Windows XP"

in Mac OS, the equivalent would be:

in /Library/Preferences/SystemConfiguration/com.apple.Boot.plist add

CODE

<key>Kernel Flags</key>
<string>-legacy</string>

it is not coming 100% close to Linux' acpi_os_name="Microsoft Windows XP", but still, have you guys ever tried that .. ?

[geiman]

hmmm, no I haven't ever tried that. I didn't even ever think of doing that with Linux, that would be great if I could get my nix os back!

EDIT: After a second look, I think I have tried what you suggested. Correct me if I'm wrong, but wouldn't that be the same as booting with the -legacy flag after pressing F8 at boot? If so, then yes, I have tried that and it makes no difference.

[BugsB]

yes, same as booting with the -legacy flag after pressing F8 at boot

[mentorek]

in grub.conf, you would just add this parameter to the kernel line, like this:

CODE

acpi_os_name="Microsoft Windows XP"

in Mac OS, the equivalent would be:

in /Library/Preferences/SystemConfiguration/com.apple.Boot.plist add

CODE

<key>Kernel Flags</key>
  <string>-legacy</string>

it is not coming 100% close to Linux' acpi_os_name="Microsoft Windows XP", but still, have you guys ever tried that .. ?

No, I've never tried -legacy option.

By the way, I've tried EFI Strings, more than 3 versions of BIOS, different Azalia kexts, HDAEnabler.kext. Everything failing. The only thing that was very close to getting sound - as a matter of fact I've got it - was very very very silent sound with Azalia but it's not acceptable at all.

[geiman]

I've never been able to hear anything in OS X since I first decided to upgrade my bios, a decision I will kick myself for until I get a new laptop. However, in Linux I have been able to hear a very faint sound coming from the speakers, however the volume control does nothing and to hear it I have to put my ear against the speaker. I'll try the Linux hack mentioned above but I'm starting to think I have done this before, only in a different way.

[Thomahawk]

After I installed the modded BIOS (rev 1.01), i had no sound anymore in ubuntu 8.04...

I use the Conexant CX20551 (Waikiki) soundchipset. Do you also have that one, or do you use the Conexant Venice chipset...That's e relevant difference I think...

In Windows XP I had normal sound.

Thanks for your reply ;-)

[BugsB]

I don't know about Venice or Waikiki:

00:1b.0 Audio device [0403]: Intel Corporation 82801G (ICH7 Family) High Definition Audio Controller [8086:27d8] (rev 02)

Description: Conexant HD Audio output
Hardware ID: HDAUDIO\FUNC_01&VEN_14F1&DEV_5047&SUBSYS_1179FF31&REV_1000
Manufacturer ID: 1
Product ID: 100

that means: Vendor-ID: 14F1 | Device-ID: 5047

[Thomahawk]

Yes, Device-ID 5047 is Waikiki, Device-ID 5045 is Venice ;-)

Well I'm looking forward until my Notebook is finally set-up with Leopard

[BugsB]

yeah that's me, I prefer Hawaiian surf to Italian canals ..

I've never been able to hear anything in OS X since I first decided to upgrade my bios, a decision I will kick myself for until I get a new laptop.

the trick is: look in the folder from where you applied the update which overwrote the 2.4 - it should contain a bios.bak. That bios.bak contains your entire BIOS from a to z, whereas a BIOS update from Toshiba only contains those parts of the BIOS that get updated (or deleted!) from that particular BIOS update. That's why all these people (look in the Linux forums, there are a lot) who, after they lost sound when updatiing beyond 2.4, never got their sound back by applying the 2.4 from Toshiba. The BIOSes beyond 2.4 OVERWROTE parts of the DSDT, if not even all of it.

Here's a quote from Master Justin:

Keep in mind that the DSDT for each system is totally unique, so it is best to decompile the DSDT from your laptop and then diff it against a repaired DSDT from a similar model. I found that Toshiba really varies their hardware even in the same product line though, so comparing two DSDT's may not be too useful. i.e. one Satellite P100 is not necessarily anything like another P100.

That's what we are experiencing here right now.

I have no idea why Justin's DSDT which came from a Toshiba Satellite PSPADU-033026 works on my PSPA3E-00R00DGR (P10x is of no importance because P100 in Europe = P105 in US ..). Looking at all you poor guys here it was pure luck that the modded DSDT from his works on mine. If you wanna know why look here ....

Mentorek is closest, because he had sound in Linux with a modded DSDT. Jusin followed this path here to get sound in Linux on his machine. This would have to be done prior to trying and get sound in Mac OS, and as you can see, it is a veeery long path ...

Still, I think if some of the values in Justin's DSDT that were modded from * (wildcard) to fixed values would be switched back to * (wildcard), we would actually increase the chance for a DSDT which is more compatible. But the testing again can only be done in Linux .

So - alltogeter bad news ..

Kabyl uploaded a dsdt.diff (modded compared to the original one). This is primarily for mentorek I would say. Mentorek, maybe you can narrow it down to the part which is responsible for the sound (and for the PCI bus ...).

Kabyl can only help you guys if you narrow down the culprit if possible all the way down to a certain value entry (like I did here). Thorrough analysis like this is paramount.

I hope I can supply my Linux codec dump as requested by you, mentorek, tomorrow.

[mentorek]

...

So if Justin is right then were {censored}ed! Maybe flashing to lower versions like 2.10 or even 1.70 would have any effect on that.
In theory, WinPhlash utility provided by Toshiba has Advanced settings where one could tweak some settings. It's possible to enable "zeroing block before writing" and do other stuff. To enable this open Winphlash.ini file and locate AdvancedSettings. Set it up to 1 and start the utility. Start the Winphlash and you should get Advanced Settings button. Maybe all of you had already tried it. I thought it's worth to mention.
BTW. I've got 2.40 BIOS updater on bootable CD. If anyone is interested I can provide you with ISO image. Also, I've found 1.70 BIOS for Toshiba P100 and I'm looking for 2.10 version. If anyone has other ideas let us know.

Cheers,
mentorek

[mentorek]

...

So after flashing both 1.70 and 2.10 BIOSes I have no sound. Well just the very very very very low.
DSDT dump from 1.70 shows that DSDT has changed slightly. Not a big change but still.

So, does any of you have a full BIOS image backed up somewhere? I mean the 2.40 and/or lower? That's the only hope I guess...

Cheers,
mentorek

P.S.
Bugs, does the slider for Master channel work for you in Audio MIDI Utility? I does not work for me (greyed out).

[BugsB]

hey mik,

master channel in Audio MIDI Utility is greyed out here, too (not movable), but sound OK, so that's not it.

I have a 2.1 BIOS.bak which I applied yesterday for a test --> no sound in OS X .. but I attached it to this posting, but like I said, don't expect anything from it.

I feel a bit sickish right now so I don't have the power for booting Linux and dumping the codec, that would be a lil bit too much for me right now, as rare as I deal with Linux heb ik geen zin voor Penguinen nou ..

I will contact you via PM right now because I want to get somebody else into the boat who, I think, has the capabilities to help us here a bit

PS: the full 2.4 BIOS.bak is unfortunately missing here, too ..

Attached File(s)

 bios_2.1.bak.zip ( 806.12K ) Number of downloads: 4

 

[geiman]

So, after breifly reading this updates, I take it things don't look good. I wish I had known about the .bak file before. I've reinstalled my os so many times since then, no chance of getting that back. Anyways, if I understand it right, If we were able to get a copy of the original bios before the DSDT became corrupt, we could theoretically flash that and have a functioning DSDT, correct? If so, I have a friend at college with the exact same laptop as mine, and he has never upgraded his bios, so there is a good chance his is still intact. Is there any chance we could get a backup of his bios to apply on ours?

Edit: Wouldn't you think we should be able to get this from Toshiba? Surely they would have a copy of the original somewhere, why couldn't they make it available to us?

[mentorek]

So, after breifly reading this updates, I take it things don't look good. I wish I had known about the .bak file before. I've reinstalled my os so many times since then, no chance of getting that back. Anyways, if I understand it right, If we were able to get a copy of the original bios before the DSDT became corrupt, we could theoretically flash that and have a functioning DSDT, correct? If so, I have a friend at college with the exact same laptop as mine, and he has never upgraded his bios, so there is a good chance his is still intact. Is there any chance we could get a backup of his bios to apply on ours?

Edit: Wouldn't you think we should be able to get this from Toshiba? Surely they would have a copy of the original somewhere, why couldn't they make it available to us?

It's possible to use .bak file with any other (Toshiba P100) laptop I would say.
Help from Toshiba - hmmm... I would not count on them but who knows. Their policy is that they do not support any other OS than the ones from MS. Well, not all of the MS OSes. Just XP, Vista and maybe 2000.
One and only hope is that one of us would have some friend at Toshiba's service and he could provide us with the solution (erasing whole BIOS chip, updating with the desired version, stuff like that).
Well, these are only speculations (about BIOS being updated with just some parts), no one is 100% sure about the update process.

[geiman]

Well once I can contact my friend from college, I will try and get a backup of his BIOS, provided it's 2.4 or earlier. Then we'll just have to figure out how to apply it, or is that possible with winphlash?