Jump to content

98 (FE), 98 SP1, 98 SE + ME SHELL32.DLL fix

MDGx

By MDGx
in Windows 9x Member Projects

 Share

Recommended Posts

Max_04

Max_04

Posted
Posted
Anonymous author's answer:
Petr,

Please try the same with the original file 4.72.3812.600 from M$! PE

Explorer also shortens the original file on save. There seems to be

something unusual in how M$ arranged the sections of this DLL.

HTH

The size change is not an issue, crashing Restorator is the issue.

I suppose that this crash is caused by improperly calculated numbers in some headers (offsets, lengths, ...), I don't know which, but if PE Explorer can rectify this problem, it should be possible to correct the origibnal file too.

Petr

There is always this problem, also in the new version in SHELL98.

Fortunately I've used french localization of glock_94 for italian localization:

http://uploaded.to/?id=a1uzkn

Without necessity to contact anonymous author.


MDGx

MDGx

Posted
  • Author
Posted

Max_04:

Anonymous author replies:

'Max_04' wrote on Jan 30 2007, 05:48 PM:

> There is always this problem, also in the new version in SHELL98.

> Fortunately I've used french localization of glock_94 for italian

> localization:

> http://uploaded.to/?id=a1uzkn

> Without necessity to contact anonymous author.

It may be a bug in Restorator, but not an error in SHELL32.DLL

4.72.3812.634. Please try to modify 4.72.3812.600 (language: English) and

see if it also crashes Restorator.

I noticed that PE Explorer or Restorator changed the DWORD at offset

0x00083098 from 0x00155000 (English) to 0x00154E00 (Italian). It needs to

be 0x00155A00 in the Italian version. This value may also be wrong in the

French version - I did not check.

I hope this helps.

Max_04

Max_04

Posted
Posted (edited)
Max_04:

Anonymous author replies:

'Max_04' wrote on Jan 30 2007, 05:48 PM:

> There is always this problem, also in the new version in SHELL98.

> Fortunately I've used french localization of glock_94 for italian

> localization:

> http://uploaded.to/?id=a1uzkn

> Without necessity to contact anonymous author.

It may be a bug in Restorator, but not an error in SHELL32.DLL

4.72.3812.634. Please try to modify 4.72.3812.600 (language: English) and

see if it also crashes Restorator.

I noticed that PE Explorer or Restorator changed the DWORD at offset

0x00083098 from 0x00155000 (English) to 0x00154E00 (Italian). It needs to

be 0x00155A00 in the Italian version. This value may also be wrong in the

French version - I did not check.

I hope this helps.

@MDGx, send this to the anonymous author:

I've tried to edit shell32.dll 4.72.3812.600 (language: English), no crash with restorator.

I've changed at offset 0x00083098, just in 0x00154E00 (Italian) 4E in 5A.

Please check me if shell32.dll now is correct:

http://uploaded.to/?id=2fk2mb

Thank you.

Ps: What is changed with this little modification?

Max

Edited by Max_04
MDGx

MDGx

Posted
  • Author
Posted

Max_04:

Reply from anonymous author:

I will look into it when I have a bit more time. I am not sure what is

causing this. The programs I tried either had a problem with the original

600 and the patched 620, 634, etc. versions (unexpected file format, C++

exception, ...) or no problem with any of the versions (600, 620, 634,

...) on save after some resource modifications.

'Max_04' wrote on Feb 5, 2007 06:11 AM:

> I've tried to edit shell32.dll 4.72.3812.600 (language: English), no

> crash with restorator.

This is odd. Thanks for trying this and letting me know.

>

> I've changed at offset 0x00083098, just in 0x00154E00 (Italian) 4E in

> 5A.

> Please check me if shell32.dll now is correct:

> http://uploaded.to/?id=2fk2mb

Please run 'MODIFYPE SHELL32.DLL -c' on the file to correct its checksum.

>

> Thank you.

>

> Ps: What is changed with this little modification?

It just points to some debug information - not needed during normal

operation.

I hope this helps.

Tihiy

Tihiy

Posted
Posted 

push  0x40
push  0x14
call  KERNEL.LOCALALLOC
mov   cx, ds <-- ds<>0
mov   [bp-4], ax
mov   [bp-2], ds
or	cx, ax <-- always <>0
jnz   ...
xor   ax, ax
jmp   ...

Holy cow, i can see that too! WE MUST FIX THIS NOW

If Anonymous patcher won't, i will.

Max_04

Max_04

Posted
Posted (edited) 
push  0x40
push  0x14
call  KERNEL.LOCALALLOC
mov   cx, ds <-- ds<>0
mov   [bp-4], ax
mov   [bp-2], ds
or	cx, ax <-- always <>0
jnz   ...
xor   ax, ax
jmp   ...

Holy cow, i can see that too! WE MUST FIX THIS NOW

If Anonymous patcher won't, i will.

If there is in shell98, please correct me this and after run 'MODIFYPE SHELL32.DLL -c' on the file to correct its checksum. :whistle:

http://www.mdgx.com/files/SHELL98I.EXE

Edited by Max_04
MDGx

MDGx

Posted
  • Author
Posted

Anonymous author replies:

Tihiy wrote on Feb 6 2007, 03:55 PM:

> Holy cow, i can see that too! WE MUST FIX THIS NOW

> If Anonymous patcher won't, i will.

I am sorry, but I have no plans to fix it because I lack the time to write

the patch program that is necessary in this case. Just patching USER.EXE

and sending it to MDGx will not do the trick here: USER.EXE needs to be

language specific to be useful, there are too many Win9x USER.EXE

hotfixes, and User/Serial # information is kept in USER.EXE. It would be

great if some other forum member could write a little program that

searches for the locations of this bug and 'or cx, ax' ---> 'or ax, ax' to

fix it.

I hope this helps.

Tihiy

Tihiy

Posted
Posted
Just patching USER.EXE

and sending it to MDGx will not do the trick here: USER.EXE needs to be

language specific to be useful, there are too many Win9x USER.EXE

hotfixes, and User/Serial # information is kept in USER.EXE.

No, that's wrong. User/Organization/Serial was in USER.EXE only in Windows <95.

And patching latest USER.EXE for win98 and winme will be enough.

However, after patching i haven't noticed any stability improvements, all my resource-draining tests still usually hang system.

So i don't think those patches will be useful. I wish i had an debug win9x version...

Rick Chauvin

Rick Chauvin

Posted
Posted (edited)

Lately I've been very often getting lots of Send Error Report prompts these days when browsing the web, and especially! when opening the History tab which opens the left window for it - and clicking any previous links of the day within that window is where moreso I get bitten with SendErrorReport prompts which aggravatingly forces me to make a play - all of which closes all browser windows of course loosing your research place, and sometimes locks everything up forcing a soft/sometimes/hard reboot...scandisk, etc....

Getting fed up with it still doing that, today I started going backwards undoing the last things I did since this started happening - but I couldn't solve it . . . until I had the thought... no way, it couldn't be the shell32.dll fix could it! ...no, please no I clamored because I want the shell32.dll fix to succeed ..but AFAICT at this moment anyway, yes it does seem to be somehow connected?

When I put the original Shell32.dll back in place I can not reproduce Send Error Report prompts as readily and it's very seldom that it ever happens as usual or so it seems anyway, but with the new Shell32.dll (latest v4.72.3812.634) in place I can readily reproduce it as indicated.

(I noticed one time that swapping the dll forward erases all current day History listings so that there tells you that it is connected internally in some way?)

It is hard to pin down helpful info for this for MDGx or Anonymous, but each time it happens the dw.log is written to, but interestingly each time it happens it lists various different browser files as the cause.

Since I do so much online research this is not something I can live with, and nOt wanting the file delete hang to pester me either so either I will revert to the v5.5 browse.dlls again (shudder) or just live with it.. ..not likely.

--Rick

oh ps, does anyone have a copy of the last shell32.dll version from anonymous that "only" has the 2-4 GB file fix in it and nothing! else - can you give a link for it please. (I knew I should of saved it but overwrote it with the new ones when they came out.

Edited by Rick Chauvin
Rick Chauvin

Rick Chauvin

Posted
Posted (edited)

Thank you much whatever420,

What are these .rc menu item mod's you have extra in the mod folder - I can see what they each are inside looking them over - what were the different things you were trying to do?

Do you have the original exe as it was offered by MDGx - I would like to have that too..

MDGx, is shell32.dll v4.72.3812.620 the last version change anonymous made for the 2-4gb fix before he started adding explorer hang fixes?

Thank you again,

Rick

Edited by Rick Chauvin
whatever420

whatever420

Posted
Posted (edited)
What are these .rc menu item mod's you have extra in the mod folder - I can see what they each are inside looking them over - what were the different things you were trying to do?

Do you have the original exe as it was offered by MDGx - I would like to have that too..

Hi Rick...

Sorry, I don't have the original EXE file... I always unpack everything so I can manually replace files...

The Shell32.dll included in the cab is the original un-modded 4.72.3812.620 dll...

The mod files are the clear shortcut icon #30 and the menu rc files are changes to the position the the "delete" command (when you right-click a file)...

I find myself deleting files when I mean to do something else (rename, etc.)... so I moved the "delete" command to a slightly different position as to avoid that...

delete01.png

Hope that makes sense... :)

Edited by whatever420
erpdude8

erpdude8

Posted
Posted
Do you have the original exe as it was offered by MDGx - I would like to have that too..

MDGx, is shell32.dll v4.72.3812.620 the last version change anonymous made for the 2-4gb fix before he started adding explorer hang fixes?

Thank you again,

Rick

the shell32.dll v4.72.3812.620 fix was pulled off MDGx's web site after version 4.72.3812.634 was posted. that was more than a month ago.

you may want to email MDGx about the older shell32.dll fix, Rick.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share
Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...