On Tue, 5 Apr 2016, Ralf Baechle wrote: > > > * Relocation is supported only by multiples of 64k bytes. This > > > eliminates the need to handle R_MIPS_LO16 relocations as the bottom > > > 16bits will remain the same at the relocated address. > > > > IIUC, that's actually better than x86, which needs to be 2MB aligned. > > On MIPS a key concern was maintaining a reasonable size for the final > kernel image. The R_MIPS_LO16 relocatio records make a significant > portion of the relocations in a relocatable .o file, so we wanted to > get rid of them. This results in a relocation granularity of 64kB. > If we were truely, truely stingy we could come up with a relocation format > to save a few more bits but I doubt that'd make any sense. Additionally, for historical reasons, with 32-bit (o32) ELF images the REL relocation format is used making borrow propagation from R_MIPS_LO16 to its corresponding R_MIPS_HI16 relocation a pain to handle. It is solvable as the static linker does handle it, in particular doing the reasonable thing for orphan relocations, but I think it's a complication worth avoiding if the cost is so little. > > > * In 64 bit kernels, relocation is supported only within the same 4Gb > > > memory segment as the kernel link address (CONFIG_PHYSICAL_START). > > > This eliminates the need to handle R_MIPS_HIGHEST and R_MIPS_HIGHER > > > relocations as the top 32bits will remain the same at the relocated > > > address. > > > > Interesting. Could the relocation code be updated in the future to > > bump the high addresses too? > > It could but yet again, the idea was to keep the size of the final > generated file under control. The R_MIPS_HIGHER and R_MIPS_HIGHEST > relocations can be discarded if we constrain the addresses to be in > a single 4GB segment. Removing this constraint would make a kernel > image much bigger so I suggested to add this restriction at least for > this initial version. For the record, with 64-bit ELF images the RELA relocation format is used, so there's no such concern about borrows as with 32-bit ones, because the whole addend is always readily available and does not have to be calculated from parts coming from different relocations. Consequently the handling of R_MIPS_HIGHER and R_MIPS_HIGHEST (and also R_MIPS_HI16 and R_MIPS_LO16) relocations in 64-bit ELF images is straightforward if we decided to include them. I suspect extending the handling to R_MIPS_HIGHER only will suffice all use cases for the foreseeable future as I don't expect MIPS systems with more than 256TiB of RAM to appear anytime soon. FWIW, Maciej
