[WIP] Windows Vista Extended Kernel

[ilwalaihr]

9 hours ago, asdf2345 said:

I think 99% of us are using the dlls given to us by @win32, so just ask.

Also, about Ryzen, what BIOS is your board running?

On the BIOS it says:

UEFI Version: B450M-HDV R4.0 P3.70.

Is this what you were asking for? 

[asdf2345]

1 hour ago, ilwalaihr said:

On the BIOS it says:

UEFI Version: B450M-HDV R4.0 P3.70.

Is this what you were asking for? 

Alright, this rules out my issues being AGESA related.

What CPU do you have?

Also, don't want to be mean, but VRM wise, you own one of the worst consumer AM4 boards to exist.

Edited August 24, 2020 by asdf2345

[ilwalaihr]

1 hour ago, asdf2345 said:

Alright, this rules out my issues being AGESA related.

What CPU do you have?

Also, don't want to be mean, but VRM wise, you own one of the worst consumer AM4 boards to exist.

It's a Ryzen 5 2600.

I knew, this motherboard is cheap, but it's better than an A320M I guess. I didn't knew that it was one of the worst though, but I never had problems with it, and it's my first desktop anyways, and it's a quite big upgrade coming from my laptop, so I don't really care.

Edited August 24, 2020 by ilwalaihr

[asdf2345]

On 8/24/2020 at 5:02 AM, ilwalaihr said:

It's a Ryzen 5 2600.

I knew, this motherboard is cheap, but it's better than an A320M I guess. I didn't knew that it was one of the worst though, but I never had problems with it, and it's my first desktop anyways, and it's a quite big upgrade coming from my laptop, so I don't really care.

Is Windows Vista installed in UEFI mode, or Legacy mode?

 

On 8/23/2020 at 2:33 AM, win32 said:

So it appears that all of the files that are modified to accommodate SHA-2 support are not part of the extended kernel, except for winload.exe (which is just a minor patch to allow for the patched ntoskrnl to work). So the extended kernel should not interfere with SHA-2. I believe that some extended kernel users are installing new SHA-2 signed updates.

Wrong, it also modifies kernel32.dll.

 

[ilwalaihr]

6 hours ago, asdf2345 said:

Is Windows Vista installed in UEFI mode, or Legacy mode?

I install it on a VM on a physical disk, and then boot it from my computer. 

But now I'm having a problem where it doesn't load the USB XHCI Drivers (Code 39).

[asdf2345]

Just now, ilwalaihr said:

I install it on a VM on a physical disk, and then boot it from my computer. 

But now I'm having a problem where it doesn't load the USB XHCI Drivers (Code 39).

Install EasyBCD, and tell me if it complains about you being in UEFI boot

[ilwalaihr]

4 minutes ago, asdf2345 said:

Install EasyBCD, and tell me if it complains about you being in UEFI boot

So I install it from EasyBCD or I install EasyBCD on Windows Vista? 

I don't know why but it freezes now, it gives me really quick BSODs that the monitor doesn't even have time to display, but the other installs (gave same error Code 39) didn't BSOD. 

Edited August 25, 2020 by ilwalaihr

[asdf2345]

Just now, ilwalaihr said:

So I install it from EasyBCD or I install EasyBCD on Windows Vista? 

I don't know why but it freezes now, it gives me really quick BSODs that the monitor doesn't even have time to display. 

You might be experiencing the same issues I do then.

Edit: Damn, now I really think it's an AGESA problem

Edited August 25, 2020 by asdf2345

[ilwalaihr]

2 minutes ago, asdf2345 said:

You might be experiencing the same issues I do then.

Edit: Damn, now I really think it's an AGESA problem

2 minutes ago, asdf2345 said:

 

Do you also have a Ryzen PC? 

Edited August 25, 2020 by ilwalaihr

[asdf2345]

Just now, ilwalaihr said:

Do you also have a Ryzen PC? 

Yeah, 2700X and X470 Gaming Pro Carbon. Been having the same issues for a long time

[ilwalaihr]

8 minutes ago, asdf2345 said:

Yeah, 2700X and X470 Gaming Pro Carbon. Been having the same issues for a long time

It's really strange how the USB Drivers where working, but now they don't. 

It was always BSODing though, and I couldn't even see the error because of how fast it was.

If I ever fix this I will tell you.

Edited August 25, 2020 by ilwalaihr

[asdf2345]

On 6/14/2020 at 7:15 PM, win32 said:

To prevent user confusion, I strongly recommend disregarding the tutorials and contacting me directly instead about the extended kernel. The original post appears below for prudency.

  Reveal hidden contents

For x86 kernel extension, see this post:

 

Attention Windows Vista x64 users!

Did you feel wounded by the way your OS was surpassed by NT 6.1? And felt salt being poured into those wounds when BlackWingCat implemented some of those NT 6.1+ functions in his extended kernel for NT 5.0?

Now I'm here to give Windows Vista the leg up it has always deserved! All you need is CFF Explorer and ExportTableTester. (rules prohibit the distribution of modified MS binaries)

First, enter X:\Windows\System32 (where X: is the letter of your system partition) and copy kernel32.dll (for this experiment, I'm using version 6.0.6002.19623; this method is applicable to versions through 6.0.6003.20686. users with later versions check out my second post below) to another folder. Open that copy of kernel32.dll in CFF Explorer and click "Section Headers [x]" on the left sidebar.

Keep your attention on the .text section and scroll down to the bottom of the hex representation of the section, where you will find a portion filled with zeros. Then starting at offset 000BCF50, add all of the non-zero code shown in the image below:

Save the file and exit CFF Explorer. Open ExportTableTester and drag-and-drop the file into its window. Click "Edit Exports" at the bottom of the window.

Enter each of the 000BCF** offsets referenced in the left window into the "Address of Function (Hex)" field on the right, followed by the function names associated with them below.

Reopen the file in CFF Explorer, click "Rebuilder" on the left sidebar, select the checkboxes for "Rebuild PE Header" and "Update Checksum" then click "Rebuild" before saving.

Place your modified kernel32.dll in the folder of each application that requires it. Create a file labelled "xxxx.exe.local" (where xxxx is the name of the application executable) in order to direct the executable to use the modified kernel32.dll as opposed to the one in the system32/sysWOW64 folder.

If you do the same steps as I did, you should be successful in this endeavour. As a sanity check, I opened Dependency Walker x64, checked the present kernel32.dll and it had those K32* functions previously exclusive to Windows 7 and up. What a feeling of accomplishment!

WARNING

Remember to keep the "Add New Section" checkbox enabled in ExportTableTester. If not selected, the expanded export table will overwrite some other code below it, which buggered up keyboard layout/localization settings (for example, all keyboard layouts under "English (Canada)" would function like the Canadian French keyboard).

Please test thoroughly and post in this thread about any issues that may arise. ExportTableTester, according to other users, may have issues of its own, but no other known tools will manipulate PE32+ export tables the way it does.

I plan on eventually introducing additional functions in kernel32.dll as well as in other system files to meet my first goal: running Waterfox 2020.03! In fact, all of the functions that have been added to kernel32 are the missing ones required by its xul.dll (with the exception of TryAcquireSRWLockExclusive, which forwards to ntdll.dll).

Off topic, but how do you do that reveal hidden contents thing?

[win32]

12 minutes ago, asdf2345 said:

Off topic, but how do you do that reveal hidden contents thing?

You use this code:

[spoiler]
[/spoiler]

 

[asdf2345]

55 minutes ago, win32 said:

You use this code:

[spoiler]
[/spoiler]

 

Thanks, needed to use it for this post, but I kinda messed up.

[ilwalaihr]

@asdf2345 I finally solved the USB Issue! Thanks to some other person on this forum.

Now I only need to install the latest update and I should be fine.

Edited August 26, 2020 by ilwalaihr