Mike Auty
eec68707bd
Improve the vmware layer (not sure it's finished yet though).
2016-12-14 00:28:05 +00:00
Mike Auty
2c60747355
Minor fix, small but might as well fix it.
2016-12-14 00:27:41 +00:00
Mike Auty
2c02df125a
Fix the truncation of strings
...
MHL kindly pointed out that the string wasn't truncating, and it turns
out I misread the struct documentation, which truncates or pads with
\x00s to ensure the appropriate length. We now truncate to the first
null we find.
2016-12-13 20:41:09 +00:00
Mike Auty
a32c7e27a9
Add in cryptic comments nearly lost during the vtypes purge.
2016-12-13 19:40:18 +00:00
Mike Auty
058346b74d
Ensure IntermediateSymbols can handle cross-table references.
2016-12-13 19:32:22 +00:00
Mike Auty
fa427d2a2b
Add in the starts of a vmware layer and stacker.
2016-12-13 02:21:15 +00:00
Mike Auty
ed716ca12a
Convert SymbolTableInterfaces to descend from ConfigurableInterface
...
This allows SymbolTables to hold actual requirements (which
IntermedSymbolTable already did, but without pulling in the right
interface). It means that values like kernel_virtual_offset get saved.
We still need to figure out how to allow plugins to demand optional
configuration values in other places (such as the kvo) but for now the
plugin will barf if it doesn't get it.
2016-12-12 02:08:32 +00:00
Mike Auty
d43b294f36
Ensure we store all relevant config values rather than just those stores as variables.
2016-12-12 02:07:48 +00:00
Mike Auty
eae84fd8e0
Ensure we only stack when needed
...
We shouldn't be stacking unless we're required, so now
we run after the construction phase, and run our own construction
phase is we've changed anything.
2016-12-12 02:06:54 +00:00
Mike Auty
046d1955b5
Ensure that only unfulfilled symbol requirements trigger the pdb scanner.
2016-12-12 02:04:20 +00:00
Mike Auty
e2d5f51ad5
Remove old and rotten code.
2016-12-12 00:25:03 +00:00
Mike Auty
21c4e4bdef
Refactor the segmented layer out of lime for use with other layers.
2016-12-11 19:13:35 +00:00
Mike Auty
b9dbc80baf
Improve logging of automagic actitivies.
2016-12-11 18:16:02 +00:00
Mike Auty
5d15f810e1
Tidy up the config namespace around automagic a little.
2016-12-11 13:58:17 +00:00
Mike Auty
9bb0745b83
Clean up the debug messages for pdbscanning.
2016-12-08 16:29:36 +00:00
Mike Auty
2aebb99ee6
Slightly improve the output and get rid of pointless parameters.
2016-12-08 16:24:59 +00:00
Mike Auty
8a0609e800
Add in a secondary scan to determine the kernel base in randomized kernel base systems.
2016-12-08 16:24:03 +00:00
Mike Auty
13b1126b90
Guard against no progress_callback in intel layer.
2016-12-08 16:03:22 +00:00
Mike Auty
482516d93d
Rework the kvo finding code to allow for Win10 kernel base randomization.
2016-12-08 14:56:32 +00:00
Mike Auty
555f587b15
Ensure the progress value for translation layer scanning is correct.
2016-12-08 12:08:42 +00:00
Mike Auty
631ae07a26
Insert a check to use valid kernels where a pdb signature is found in multiple locations.
2016-12-08 11:55:46 +00:00
Mike Auty
183ff67699
Attach the kernel virtual offset to intel spaces.
2016-12-07 16:00:31 +00:00
Mike Auty
1f6fcd7bad
Ensure that the value of pointers always fall within their layer.
2016-12-07 09:52:33 +00:00
Mike Auty
797e51608e
Move the masking code into the data layer.
2016-12-07 09:50:59 +00:00
Mike Auty
6cb5d06d6d
Ensure all offsets are normalized to fall within their layer's address range.
2016-12-07 00:07:57 +00:00
Mike Auty
20d0e83e1d
Ensure we don't ever round down when log-ing.
2016-12-07 00:05:47 +00:00
Mike Auty
734b9a68b8
Ensure we use math.log correctly.
...
This is unlikely to ever happen (since we use python3)
but still just as well to implement in case something else
similar comes up.
https://www.willmcgugan.com/blog/tech/post/finding-the-first-bit-set-with-python/
2016-12-06 23:40:21 +00:00
Mike Auty
40694ff26c
Add a note about precedence to the pdb loader.
2016-12-06 09:08:51 +00:00
Mike Auty
a064c76628
Ensure we initialize kvo, change the filename format and give human readable feedback on the choice of natives.
2016-12-06 02:44:50 +00:00
Mike Auty
1c7de06bd0
Fix up issue with pointer sizes. Since cls is never used we're not losing information.
2016-12-06 02:44:11 +00:00
Mike Auty
c52bf7cb6f
Pull out the appropriate natives based on the intermediate format data.
2016-12-04 23:50:13 +00:00
Mike Auty
770a8ee42a
Improve error handling to prevent endless recursion in template.__call__
2016-12-04 11:17:48 +00:00
Mike Auty
05ae4aa64a
Fix up offset for PAE.
2016-12-04 10:24:33 +00:00
Mike Auty
d8b72cfc14
Add in checks and set the appropriate kernel location for all architectures.
2016-12-04 02:17:11 +00:00
Mike Auty
a0f7c81128
Reveal the register size in intel spaces.
2016-12-04 02:14:54 +00:00
Mike Auty
4121895d4d
Ensure the kernel virtual to physical offset mapping is correct.
2016-12-04 01:21:10 +00:00
Mike Auty
e1cc55aeed
Add in wchar to our set of bases for x64 systems.
2016-12-04 01:02:17 +00:00
Mike Auty
e27a444068
Very minimally fix up the docstring for a function.
2016-12-04 01:01:46 +00:00
Mike Auty
88f1166c6d
Add in a helper function that will likely be useful in the future.
2016-12-04 01:01:13 +00:00
Mike Auty
23fd704027
Add in the kernel offset recursion code.
2016-12-04 01:00:37 +00:00
Mike Auty
9a3402ccff
Fix the incorrect signature offset returned.
2016-12-03 21:39:04 +00:00
Mike Auty
e4e563ead0
Fix up the array slicing code.
2016-12-03 00:23:00 +00:00
Mike Auty
70815b0837
Shift the location of the pdbscan prefix/suffix values.
2016-12-02 21:44:23 +00:00
Mike Auty
ed552d20e8
Add in support for lzma compressed json data.
2016-12-02 01:34:07 +00:00
Mike Auty
c86f0ddef4
Add in a very rudimentary kernel PDB finding plugin that identifies and automatically loads the appropriate kernel for the image.
2016-11-27 23:40:19 +00:00
Mike Auty
b2c3c76954
Don't forget the fixes that still need to happen.
2016-11-19 23:41:35 +00:00
Mike Auty
72259008f3
Use the physical layer and improve the output.
...
The physical layer will miss certain patterns, but is an order of
magnitude faster at scanning. The main amount of time spent in scanning
Intel spaces is counting through every page in the address space (32, 40
or 64 bits), not the actual scanning. There's no real way around this
if you want to ensure you get every chunk of virtual memory. Since the
scanner could be stopped after its first hit, this might still be
preferable, but should not be the default (particularly for an automagic
scan).
2016-11-19 23:36:48 +00:00
Mike Auty
d2fcc1d4b0
Add in optimization technique to reduce the number of pages that are scanned more than once.
2016-11-19 23:34:08 +00:00
Mike Auty
c9abd290ec
Remove an unnecessary test from the intel scan loop.
2016-11-19 18:05:41 +00:00
Mike Auty
38b1c61acc
Remove debugging values from the pdb scanner.
2016-11-18 17:18:50 +00:00
Mike Auty