J. Mayer wrote: [snip] > The thing I see is different is that the n32 ABI redefines elf_greg_t > and elf_caddr_t as 32 bits. Maybe I missed something but those types > seem not to be used by the ELF loader (or maybe I should look in a more > recent kernel ;-) ). > Then, I have seen no apparent issue with the patch and I'm quite sure > that, even if it's not correct for some specific ABI, it would bring no > regression.
Agreed. Thiemo