Supermium

[mjd79]

5 hours ago, Dave-H said:

The error I'm actually getting on XP is -

"Entry Point Not Found : The procedure entry point InitializeCriticalSectionEx could not be located in the dynamic link library KERNEL32.dll."

In CFF Explorer, replace the kernel32.dll import with pwrp_k32.dll and copy this file from your Supermium installation. The screenshot was taken on XP RTM

 

Edited Monday at 11:49 PM by mjd79

[VistaLover]

5 hours ago, Dave-H said:

The error I'm actually getting on XP is -

"Entry Point Not Found : The procedure entry point InitializeCriticalSectionEx could not be located in the dynamic link library KERNEL32.dll."

Actually, "supermium_uao.exe" won't launch under XP SP3 (32-bit) because of two missing functions on XP's kernel32 system file: 

# Report By YY.Depends.Analyzer (Target:5.1.2600-x86)
## kernel32.dll
* [ ] InitializeCriticalSectionEx
  - Supported OS: 6.0.6000, 6.1.7600, 6.2.9200, 6.3.9600, 10.0.10240
  - Ref Module: supermium_uao.exe
* [ ] LCMapStringEx
  - Supported OS: 6.0.6000, 6.1.7600, 6.2.9200, 6.3.9600, 10.0.10240
  - Ref Module: supermium_uao.exe

Above is a log acquired via the YY-Thunks CLI tool; the command I used was: 

YY.Depends.Analyzer "supermium_uao.exe" /IgnoreReady /ReportView:CheckBox /Target:5.1.2600

As @mjd79 suggested (), if you redirect the kernel32.dll function calls to Supermium's wrapper DLL "pwrp_k32.dll", you'll be able to make the executable XP-compatible  ...

[user57]

8 hours ago, mjd79 said:

In CFF Explorer, replace the kernel32.dll import with pwrp_k32.dll and copy this file from your Supermium installation. The screenshot was taken on XP RTM

 

that is a kernel extender

it seems to be a overkill just for file read tool - why it cant make a simple text string with normal code

it might have been compiled with a newer version of visual studio 2019

 

i thought i might talked about this already

https://msfn.org/board/topic/182888-how-vs-makes-working-code-still-incompatible/

the vistual studio 2019 v16.1 to v16.7 dont have that problem - all after 16.8 - 16.11 have that trick inside

it is a hidden .obj file, where in the c-runtime code/ucrt/std depening on how we call this is forced to being used

you can see the header and cpp files - but if you edit them it never reads that code - instead it always force to use that .obj file - what is precompiled with that code that use vista+ functions

as said in that article that compiler makes a code that runs without your code and it runs before your programm is starting up (and there are these missing functions - even if they are not used in your code)

[Dave-H]

11 hours ago, mjd79 said:

In CFF Explorer, replace the kernel32.dll import with pwrp_k32.dll and copy this file from your Supermium installation. The screenshot was taken on XP RTM

 

Thanks, everyone. Yes, that modification makes the file work fine on XP.

[mjd79]

7 hours ago, user57 said:

that is a kernel extender

it seems to be a overkill just for file read tool - why it cant make a simple text string with normal code

it might have been compiled with a newer version of visual studio 2019

 

i thought i might talked about this already

https://msfn.org/board/topic/182888-how-vs-makes-working-code-still-incompatible/

the vistual studio 2019 v16.1 to v16.7 dont have that problem - all after 16.8 - 16.11 have that trick inside

it is a hidden .obj file, where in the c-runtime code/ucrt/std depening on how we call this is forced to being used

you can see the header and cpp files - but if you edit them it never reads that code - instead it always force to use that .obj file - what is precompiled with that code that use vista+ functions

as said in that article that compiler makes a code that runs without your code and it runs before your programm is starting up (and there are these missing functions - even if they are not used in your code)

Of course. My suggestion was completely automatic, because I don't use Supermium and wanted to test this tool the easy way. I'm busy porting Chromium for Windows 7 to Windows 2000 SP4 without KernelEx

 

Edited Tuesday at 03:42 PM by mjd79

[VistaLover]

3 hours ago, Dave-H said:

Yes, that modification makes the file work fine on XP.

... It would appear that Shane is still monitoring this thread  ; inside the Supermium-v144-r4_01 release notes, 

https://github.com/win32ss/supermium/releases/tag/v144-r4_01

this shows up: 

Quote

The UAO application should also now work on Windows XP.

https://github.com/win32ss/supermium/releases/download/v144-r4_01/supermium_uao.zip

Edit: Rather the fix was a consequence of the discussion in: 

https://github.com/win32ss/supermium/issues/1907

 ...

Edited Tuesday at 02:56 PM by VistaLover
Added related GH issue

[Dave-H]

Yes indeed!
He's done the same edit that I did to the file and bundled a copy of pwrp_k32.dll in the zip file with it.

[modnar]

Looks like there is a new issue to patch: "Google is aware that an exploit for CVE-2026-11645 exists in the wild,...".

[NotHereToPlayGames]

What does "exists in the wild" *REALLY* mean?  ie, does it really affect v144?  Or does it just affect those that "thought" they were safe by always running the most recent?

[modnar]

17 hours ago, NotHereToPlayGames said:

What does "exists in the wild" *REALLY* mean?  ie, does it really affect v144?  Or does it just affect those that "thought" they were safe by always running the most recent?

It means the exploit exists in the wilderness that google created with their crap, I mean chrome. V 149 (their latest) was patched.

[NotHereToPlayGames]

Okay, but that kind of misses my question.

IF this only affects v149, then IT DOES *NOT* EFFECT Supermium.

ie, "the sky isn't falling"

 

edit: ie, the "stupidly aggressive" update schedule (that both Chrome and Firefox do!) **CAUSED** the "exploit"

so *ONLY* people that "update every d@mn day" (exaggerating for effect) were even effected

Edited 22 hours ago by NotHereToPlayGames

[user57]

from what i remember "in the wild" was a description for commercial buiness style guys
useally they have something they dont like/want (like XP)
also useally the meaning exits that only they are professionals and only their work and what they want should exits (again xp is such an example, they do not want to have it, the title in the c-runtime question was changed from "getting rid of xp" to "Remove XP (and Server 2003) support") - that shows what they want - they could have keept it - but they dont - "they are the professionals - they "should" decide how this is going on"
so now if something was around they called it exits "somewhere in the wild" - what they cant get rid of so easy, but they useally always try

you are leaving them a possible option to remove your programm "from the wild" by tolerating Not_A Brand with "version 8" - this one should be changeable - or at some point they will use this one

 

in the past i pointed out to build in the the user agent (build in no plugins) would be a good idea - now that theory is proofen to be right

 

we are not a company, we are not a big concern, we are not microsoft

i think modders, retro, maybe open source stuff rather would describe this kind of software around here

but the stuff around here often actually is good, its not like that here are dump people or something rather its the opposite