You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
syscall: NtResumeThread pid=4172 tid=8264
make_writable: pc 0x00000000154c3000-0x00000000154ed000, currently r--- committed
make_unwritable: pc 0x00000000154c3000-0x00000000154ed000, currently rw-- committed
non-excluded, non-preinjected child should be injected
injecting into child process
is_first_thread_in_new_process: pid=0x104c vs me=0x1db8, xbx=0x000000007efde000 vs peb=0x000000007efdf000
Not first thread so not setting DR env vars in pid=0x104c
on vista+ the wow64 init sequence changed: kernel maps in 32-bit ntdll (xref issue #803 ), and
apparently has the first thread arg point at the 32-bit PEB instead of the
64-bit PEB. PEB from ProcessBasicInformation for wow64 process is 64-bit
PEB. this breaks the is_first_thread_in_new_process() heuristic. I may just
assume the PEB's are on subsequent pages for now (never seen otherwise).
From bruen...@google.com on June 21, 2012 12:11:00
using earliest injection on 32-bit child:
syscall: NtResumeThread pid=4172 tid=8264
make_writable: pc 0x00000000154c3000-0x00000000154ed000, currently r--- committed
make_unwritable: pc 0x00000000154c3000-0x00000000154ed000, currently rw-- committed
non-excluded, non-preinjected child should be injected
injecting into child process
is_first_thread_in_new_process: pid=0x104c vs me=0x1db8, xbx=0x000000007efde000 vs peb=0x000000007efdf000
Not first thread so not setting DR env vars in pid=0x104c
on vista+ the wow64 init sequence changed: kernel maps in 32-bit ntdll (xref issue #803 ), and
apparently has the first thread arg point at the 32-bit PEB instead of the
64-bit PEB. PEB from ProcessBasicInformation for wow64 process is 64-bit
PEB. this breaks the is_first_thread_in_new_process() heuristic. I may just
assume the PEB's are on subsequent pages for now (never seen otherwise).
xref comment at http://msdn.microsoft.com/en-us/library/windows/desktop/ms684280(v=vs.85).aspx xref PR 215400: win32 start address wrong
no, there's no issue w/ using the 32-bit vs 64-bit thread start arg: it is 64-bit in all other respects
Original issue: http://code.google.com/p/dynamorio/issues/detail?id=816
The text was updated successfully, but these errors were encountered: