Thanks George,
*With virtual environments it depends on how the emulated BIOS constructs
the virtual memory map. *
On multiple PCs running Vmware workstation, my gap was always
between 0xC0000000 - 0x100000000.
Also, in vmware-tools for Linux (C:\Program Files (x86)\VMware\VMware
Workstation\linux.iso), the script vmware-tools-distrib\bin\
vmware-config-tools.pl contains these lines:
if (is64BitKernel()) {
$gSystem{'page_offset'} = '0000010000000000';
} else {
$gSystem{'page_offset'} = 'C0000000';
}
Maybe vmware emulated BIOS always reserve the range 0xC0000000 -
0x100000000?
Sebastien
On Wed, Aug 7, 2013 at 1:00 PM, George M. Garner Jr. <
ggarner_online(a)gmgsystemsinc.com> wrote:
Sébastien,
It depends on the hardware installed on your PC,
but most of the time
(on my PCs), the range to pad was between 0xC0000000 - 0x100000000 <
For real hardware the physical memory map is programmed into the memory
controller by the BIOS. It depends on the physical hardware, OEM design
choices and BIOS manufacturer. The operating system also may (usually
does) remove some pages from the physical memory map. The most notable
example is so-called "unmanaged" memory on x86 Windows systems with more
than 4 GiB of RAM. But virtually every PC has at least some reserved pages.
With virtual environments it depends on how the emulated BIOS constructs
the virtual memory map. Hyper-V has a lot of synthetic device drivers that
do not require PCI physical address space. The reserved space just below 4
GiB is greatly reduced.
Intel and AMD systems used to be fairly uniform in their memory map, but
that is no longer the case. We are encountering an increasing number of
systems that reclaim physical address space above 0xC0000000 or that have
multiple small memory holes between 0xB0000000 and 0xC0000000. What is
worse, in some cases the processor will generate a bugcheck if you try to
read from these memory "holes."
The long and the short of it is that physical memory acquisition is
becoming increasingly complex. There is no simple one-size-fits-all
solution.
Regards,
George.
______________________________**_________________
Vol-users mailing list
Vol-users(a)volatilityfoundation.org
http://lists.volatilesystems.**com/mailman/listinfo/vol-users<http://lis…