[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference decwet::nt-developers

Title:MS Windows NT Developers
Notice:See note 1222 for MS bug reporting info
Moderator:TARKIN::LINEIBER
Created:Mon Nov 11 1991
Last Modified:Tue Jun 03 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:3247
Total number of notes:15633

3222.0. "Problems with EngFindImageProcAddress on Alpha/NT 4.0" by KAMPUS::ALTENHOFEN (Michael Altenhofen) Fri Apr 11 1997 07:04

    Hi there,
    
    We're writing a pseudo display driver for Windows NT 4.0 to intercept
    calls to the original display driver.
    Our pseudo driver is loaded instead of the original display driver, 
    installs it's functions and afterwards loads and starts the original
    display driver.
    
    To do this we use EngLoadImage to load the original driver and
    afterwards call EngFindImageProcAddress to get a pointer to the
    DrvEnableDriver function and at last the function is called.
    
    On intel x86 systems this works fine, but on a alpha system the 
    EngFindImageProcAddress function call crashes (the EngLoadImage 
    seems to work). 
    
    The error code for the crash is:
    *** Fatal System Error: 0x0000003B
    (0x00000000,0x00000000,0x00000000,0x00000000)
    
    To test it we also loaded "ntdll.dll" and tried to get function adresses
    with EngFindImageProcAddress and it worked on both intel and alpha
    systems.
    
    It seems that the display driver is no longer a "valid" DLL under NT
    4.0. We used dumpbin to check which functions are exported by
    the display driver, but it said that no functions at all are exported
    (both on intel and alpha). So is it correct that it is no longer a
    "vaild" DLL, and if so why does the intel version still work with
    EngFindImageProcAddress ?
    
    Is the code to get function addresses from a display driver not present
    on an alpha (if so how can we do it ?), or is there an bug in the
    EngFindImageProcAddress on alpha (is there a work around ?) ?
    
    If anybody of you can shed some light on this please send an e-mail to
    
    	altenhof@kar.dec.com
    
    and
    
    	krebs@kar.dec.com
    
    Thanks for your help
    
    Michael Altenhofen
    
    CEC Karlsruhe
    European Applied Research Center
    
T.RTitleUserPersonal
Name
DateLines