Hi George,
Yeah I forgot that pae images do not have to have page tables
aligned to page size. You are correct.
It would be interesting to know if the original address range was
captured in the image in the first place. As you pointed out that
address is toward the top end of the address range on 32 bit machines.
I know that win32dd has a bug where it would truncate the image short
(thus missing the top of the address space) but I thought this was
fixed in dumpit (see Issue 198
https://code.google.com/p/volatility/issues/detail?id=198)
Is the address 3476263616 (0xcf3392c0) within the image? Is the image
at least 3.4gb large? Do both images produced from the different
tools have the same size?
Michael.
On 30 October 2012 22:49, George M. Garner Jr.
<ggarner_online(a)gmgsystemsinc.com> wrote:
Rob,
Glad that you solved your problem. Just to clarify something, if PAE is
enabled the DirectoryTableBase member of the _KPROCESS structure contains
the physical address of the PDP which is aligned along 20h byte boundaries
and not page boundaries. See e.g.
PROCESS 85ba45e8 SessionId: 0 Cid: 090c Peb: 7ffd8000 ParentCid: 0228
DirBase: 7ef6c400 ObjectTable: 98a20180 HandleCount: 350.
Image: svchost.exe
0: kd> dq /p /c2 7ef6c400 L4
7ef6c400 00000000`26f2e801 00000000`086bf801
7ef6c410 00000000`27100801 00000000`26d69801
The physical address cf3392c0h is aligned for a possible x86 PDP, except
that it is within a physical address range that typically is reserved for
use by the PCI bus (c0000000h-100000000h). However, some recent AMD systems
have begun using some addresses above c0000000h for RAM physical addresses.
So without knowing more about the design of the system and motherboard
chipset I cannot say whether or not that is a valid PDP physical address
(assuming that you have PAE enabled, which is typical for server systems).
I suppose that you rebooted the system in between samples so that we cannot
compare the DTB values from the two memory "dumps" directly?
Regards,
George.
_______________________________________________
Vol-users mailing list
Vol-users(a)volatilityfoundation.org
http://lists.volatilityfoundation.org/mailman/listinfo/vol-users