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

Conference vaxaxp::vmsnotes

Title:VAX and Alpha VMS
Notice:This is a new VMSnotes, please read note 2.1
Moderator:VAXAXP::BERNARDO
Created:Wed Jan 22 1997
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:703
Total number of notes:3722

550.0. "?How to map process' VA to PA?" by RICKS::OPP () Tue Apr 29 1997 22:43

    	The $LCKPAG system service returns the virtual address range 
    that was locked into the working set if the function was successful.
    Is there a system service available that would convert the virtual
    address of each page that was locked to it's physical address?  
    
    
         As I recall, this can be done readily via SDA.  However, an 
    enhancement to a currently existing memory test program has been re-
    quested to obtain the physical addresses, in order to map the working 
    set to the SIMMs.  (There are a number of address/data translations 
    that must be performed to successfully do this.  From the program-
    ming perspective, obtaining the physical addresses of the pages 
    locked into the process' working set is one that seems viable to 
    add to the test program.)   Any suggestions would be appreciated.
    
    Greg
T.RTitleUserPersonal
Name
DateLines
550.1COMEUP::SIMMONDSloose canonTue Apr 29 1997 23:5310
    Re: .0
    
    Assuming you're talking Alpha, how about mmg$pteref() and then
    ioc$svapte_to_pa() ?  ('device driver' or similar context required
    for these, because of Spinlocks, etc.)
    
    I haven't seen a supported system service which would do all this for
    you..
    
    John.
550.2EVMS::KUEHNELAndy K�hnelWed Apr 30 1997 10:174
    Since the page is already locked in memory and in your working set,
    you can simply use the pte_va macro from VMS_MACROS.H to get the
    address of the PTE mapping the address.  The high order longword of the
    PTE contains the PFN.
550.3Danke schonGREGOR::OPPWed Apr 30 1997 13:596
    RE: .1 & .2
    
    	Thank you for the suggestions.  I'll check them out.
    
    GLO