bug-binutils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Bug ld/31179] RISC-V: The SET/ADD/SUB fix breaks ABI compatibility with


From: palmer at gcc dot gnu.org
Subject: [Bug ld/31179] RISC-V: The SET/ADD/SUB fix breaks ABI compatibility with 2.41 objects
Date: Fri, 22 Dec 2023 16:25:36 +0000

https://sourceware.org/bugzilla/show_bug.cgi?id=31179

--- Comment #9 from Palmer Dabbelt <palmer at gcc dot gnu.org> ---
(In reply to Nelson Chu from comment #8)
> Created attachment 15270 [details]
> proposed solution with the tag to keep compatible
> 
> Updated to have a tag, this patch should be applied after the proposed v1
> patch.
> 
> The object which marked by the tag means it won't have any non-zero ddend in
> SUB_ULEB128 from .uleb128 directives, so we don't need to ignore them to get
> the
> correct calculation in linker.
> 
> Since the tag is non-standard in riscv psabi, so need use
> elf_other_obj_attributes_proc rather than elf_known_obj_attributes_proc api
> to access.

Thanks Nelson, that was super fast ;)

So I think this is a roughly workable solution (maybe we should cache those tag
lookups for performance, not sure if it matters).  It's going to be hard to
tell for sure without some distro testing, though.

Does this pass the toolchain regressions?  If so, I think we should poke the
distro folks (many of whom I guess who are here) to see if it's actually
resolving their issues.

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


reply via email to

[Prev in Thread] Current Thread [Next in Thread]