Subject:
Re: reverse address translation
From:
Chip Coldwell <coldwell@physics.harvard.edu>
Date:
Mon, 20 Dec 2004 12:32:07 -0500
Newsgroups:
comp.os.vms

On Mon, 20 Dec 2004, Bob Koehler wrote:

> In article <Pine.LNX.4.61.0412191718480.10332@frank.harvard.edu>, Chip Coldwell <coldwell@physics.harvard.edu> writes:
>
>>
>> Is there any way to get the virtual address that corresponds to a given
>> PFN?
>
>
>   There may be many, or none.
>
>   The technique goes something like this:
>
>   1) buy the internals manual
>   2) search the memory map for the PFN and report
>   3) if the page is shareable, repeat 2) for all processes
>
>   Oh, you can use SDA to get at this data, just plan to take lots of
>   time.


Really, I phrased that question very badly.  What I've been trying to do is to take a peek inside the memory set aside for console use.  I can find out which physical pagess it's using with the $GETSYI system service (with the SYI$_PFN_MEMORY_MAP flag).  It turns out to be possible to map those pages into process address space using $CRMPSC (with the SEC$M_PFNMAP flag).  I was surprised by that, since the physical pages are flagged to be for use by the console, not openvms.  But nothing prevented me from doing it.

I've got the internals manual, which is how I figured out to try the procedure above.

I'm still trying to figure out where this SRM BIOS emulator's emulated x86 memory lives.  Definitely would like to hear from someone who's worked on it.

Chip

-- 
Charles M. "Chip" Coldwell
System Administrator
Harvard Physics Department
617-495-3388