2022-07-28 12:36:07

by Huacai Chen

[permalink] [raw]
Subject: Re: [PATCH v2 0/4] LoongArch: Support new relocation types

Hi, Arnd,


On Thu, Jul 28, 2022 at 7:49 PM Xi Ruoyao <[email protected]> wrote:
>
> The version 2.00 of LoongArch ELF ABI specification introduced new
> relocation types, and the development tree of Binutils and GCC has
> started to use them. If the kernel is built with the latest snapshot of
> Binutils or GCC, it will fail to load the modules because of unrecognized
> relocation types in modules.
>
> Add support for GOT and new relocation types for the module loader, so
> the kernel (with modules) can be built with the "normal" code model and
> function properly.
>
> Tested by building the kernel with Binutils & GCC master branch,
> running the kernel with 35 in-tree modules loaded, and loading one
> module with 20 GOT loads (both old SOP_PUSH_GPREL and new
> GOT_PC_HI20/GOT_PC_LO12 relocations tested, and loaded addresses
> verified by comparing with /proc/kallsyms).
>
> Link: https://github.com/loongson/LoongArch-Documentation/pull/57
> Link: https://gcc.gnu.org/r13-1834
> Link: https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=f09482a
Could you update the cross-compilers here [1] to make this series be
possible upstream in the 5.20 cycle? Or we can only do that until new
binutils/gcc be released?

[1] https://mirrors.edge.kernel.org/pub/tools/crosstool/

Huacai
>
> Changes from v1 to v2:
>
> - Fix a stupid programming error (confusion between the number of PLT
> entries and the number of GOT entries). (Bug spotted by Youling).
> - Synthesize the _GLOBAL_OFFSET_TABLE_ symbol with module.lds, instead
> of faking it at runtime. The 3rd patch from V1 is now merged into
> the 1st patch because it would be a one-line change. (Suggested by
> Jinyang).
> - Keep reloc_rela_handlers[] ordered by the relocation type ID.
> (Suggested by Youling).
> - Remove -fplt along with -Wa,-mla-* options because it's the default.
> (Suggested by Youling).
>
> Xi Ruoyao (4):
> LoongArch: Add section of GOT for kernel module
> LoongArch: Support R_LARCH_SOP_PUSH_GPREL relocation type in kernel
> module
> LoongArch: Stop using undocumented assembler options
> LoongArch: Support modules with new relocation types
>
> arch/loongarch/Makefile | 4 --
> arch/loongarch/include/asm/elf.h | 37 ++++++++++
> arch/loongarch/include/asm/module.h | 23 ++++++
> arch/loongarch/include/asm/module.lds.h | 1 +
> arch/loongarch/kernel/head.S | 10 +--
> arch/loongarch/kernel/module-sections.c | 51 ++++++++++++--
> arch/loongarch/kernel/module.c | 94 +++++++++++++++++++++++++
> 7 files changed, 207 insertions(+), 13 deletions(-)
>
> --
> 2.37.0
>


2022-07-28 14:57:47

by Xi Ruoyao

[permalink] [raw]
Subject: Re: [PATCH v2 0/4] LoongArch: Support new relocation types

On Thu, 2022-07-28 at 20:33 +0800, Huacai Chen wrote:
> Hi, Arnd,

/* snip */

> Could you update the cross-compilers here [1] to make this series be
> possible upstream in the 5.20 cycle? Or we can only do that until new
> binutils/gcc be released?

Hi Huacai and Arnd,

The old toolchain will use

R_LARCH_PUSH_PCREL _GLOBAL_OFFSET_TABLE_
R_LARCH_PUSH_GPREL symbol
R_LARCH_ADD

to perform a PC-relative addressing for a GOT entry, and the new
toolchain will use R_LARCH_GOT_PC_{HI20,LO12}. Both way are supported
by this series.

I've tested the V3 patch (V2 + suggestions from Youling, I'll sent it
later after rewording the third patch) with both my system toolchain (no
new reloc support) and a toolchain built from binutils & gcc master.
Both of the build works fine.

So a toolchain update seems not immediately required.
--
Xi Ruoyao <[email protected]>
School of Aerospace Science and Technology, Xidian University