[PATCH v3 0/8] bfd: harden in-place relaxation byte deletion against size underflow

Jan Beulich jbeulich@suse.com
Fri Nov 13 07:52:41 GMT 2026


On 13.11.2026 08:47, Harshit Kumar wrote:
> Ping for the rest of v3.
> 
> Jan, thank you for committing 6/8 (rl78: return false from
> relax_section error path) as b04483c4ba6.
> 
> The remaining seven patches (1-5, 7 and 8) still apply cleanly to
> current master (b6272d0e9071) without 6/8.  v3 addresses your v2
> comments on 1/8 and 2/8: the helper parameters are now 'offs' and
> 'limit', abfd and sec are pointer-to-const, the diagnostic reports
> offs, count, limit and the section size, the redundant cast in
> elf32-v850.c is gone, and the target tests share ld_relax_underflow_test.
> 
> Are 1/8 and 2/8 OK now?

Well, it's getting close, but not quite there: Please wait for at least a
week before pinging. I, for one, have usually only Fridays to look after
binutils stuff. Hence you pinged slightly too early.

>  The backend patches depend on 1/8:
> 
>   3/8 avr      - Denis
>   4/8 m68hc11  - Stephane
>   5/8 rx, 7/8 rl78
>   8/8 riscv    - Palmer, Kito
> 
> An ack from the respective maintainers would help.

Yes, albeit I'm unsure upon what basis you picked the two (out of five)
RISC-V maintainers.

Jan


More information about the Binutils mailing list