[Nasm-bugs] [Bug 3392533] Considerations for segment support in ELF

[noreply-nasm at gorcunov.org]

https://bugzilla.nasm.us/show_bug.cgi?id=3392533

--- Comment #25 from TK Chia <u1049321969 at caramail.com> ---
Hello hpa, hello stsp,

I recently took the chance to make my existing relocation scheme much less
hacky.  The binutils-ia16 linker can now directly process R_386_SEGMENT16
relocations and will magically create the needed MZ relocations (unlike before,
https://bugzilla.nasm.us/show_bug.cgi?id=3392533#c11).  This I achieved by
defining a new "linker emulation" to specially process the relocations. 
(Binutils defines special "linker emulations" for targets such as ARM, which
are used to create ARM-specific thunk routines and do other deep magic.)

While at it, I also added hjl's initial implementation of your proposed
R_386_{SEG16, SUB16, SUB32} relocations, and made some further changes to get
them semi-working.  (I renamed R_386_SEG16 to R_386_SEG16X, to reduce confusion
with the existing R_386_SEGMENT16.)

The code is under a branch in my binutils-ia16 Git repository
(https://github.com/tkchia/binutils-ia16/tree/tkchia/saner-mz-relocs).

Currently, my existing relocation ABI is not exactly compatible with the new
proposed relocation ABI, so there probably needs to be a way to bridge or unify
the two...

Thank you!

-- 
You are receiving this mail because:
You are watching all bug changes.
You are on the CC list for the bug.