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
The global only tells us where g_actor_name lives, but we want to also check whether the (non-c++) constructor call is identical.
So I think it also needs a // CONSTRUCTOR: LEGO1 ... annotation.
So is this a combination of decomplint and reccmp?
Do PDB's have information about this?
In LEGO1, there are 7 of these pointers between xc_a and xc_z in .data. The first sets an exit callback with atexit and calls KERNEL32.DLL::SetUnhandledExceptionFilter. No name in the PDB.
The other 6 are thunks that point to the constructor functions. All 12 of those appear to correspond to the S_LPROC32 symbols in cvdump output.
The 6 constructor functions have a corresponding line reference in the LINES section, so it would be possible to add a new CONSTRUCTOR marker type that could run alongside variables and match these up.
The text was updated successfully, but these errors were encountered:
Back in November, @madebr asked:
And:
In
LEGO1
, there are 7 of these pointers betweenxc_a
andxc_z
in.data
. The first sets an exit callback withatexit
and callsKERNEL32.DLL::SetUnhandledExceptionFilter
. No name in the PDB.The other 6 are thunks that point to the constructor functions. All 12 of those appear to correspond to the
S_LPROC32
symbols in cvdump output.The 6 constructor functions have a corresponding line reference in the
LINES
section, so it would be possible to add a newCONSTRUCTOR
marker type that could run alongside variables and match these up.The text was updated successfully, but these errors were encountered: