https://sourceware.org/bugzilla/show_bug.cgi?id=30390
--- Comment #3 from Nick Clifton <nickc at redhat dot com> --- Hi Pali, I must admit that I am reluctant to pursue this feature. This stub is only used when running an executable in a DOS environment yes ? So its utility in modern Windows based systems seems to be quite small, and implementing it is likely to introduce new bugs into the linker. For the record the function where the hard-coded message is held is bfd/peicode.h:pe_mkobject(). The message is stored in a 64 byte array in the pe_data_type structure. Adding an option to store a custom sequence of 64 bytes into this structure would not be too hard, but I imagine that what you really want is to be able to store a stub of an arbitrary length. This would mean putting in a lot more changes, and would be much more likely to break things. This is why I am reluctant to have a go at this. Cheers Nick -- You are receiving this mail because: You are on the CC list for the bug.
