Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <a30e39a1b41dc63fff32b8d350fecb92dfe27d09.camel@xry111.site>
Subject: Re: [PATCH v3 0/4] LoongArch: Support new relocation types
From:   Xi Ruoyao <xry111@xry111.site>
To:     Youling Tang <tangyouling@loongson.cn>, loongarch@lists.linux.dev
Cc:     linux-kernel@vger.kernel.org, WANG Xuerui <kernel@xen0n.name>,
        Huacai Chen <chenhuacai@kernel.org>,
        Jinyang He <hejinyang@loongson.cn>
Date:   Fri, 29 Jul 2022 10:29:53 +0800
In-Reply-To: <879dc428-0407-3280-617d-840352aa6d08@loongson.cn>
References: <bb7824d39a694b13069718c2b8193379f79229b4.camel@xry111.site>
         <879dc428-0407-3280-617d-840352aa6d08@loongson.cn>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.44.3 
MIME-Version: 1.0
Precedence: bulk

On Fri, 2022-07-29 at 10:23 +0800, Youling Tang wrote:
> Hi, Ruoyao
>=20
> On 07/29/2022 10:07 AM, Xi Ruoyao 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.=C2=A0 If the kernel is built with the latest snaps=
hot of
> > Binutils or GCC, it will fail to load the modules because of unrecogniz=
ed
> > relocation types in modules.
> >=20
> > 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.
> >=20
> > This series does not break the compatibility with old toolchain using
> > stack-based relocation types, so with the patches applied the kernel ca=
n
> > be be built with both old and new toolchains.
> >=20
> > Tested by building the kernel with both Binutils & GCC master branch an=
d
> > my system Binutils & GCC (without new relocation type support), running
> > both the builds with 35 in-tree modules loaded, and loading one module
> > with 20 GOT loads (loaded addresses verified by comparing with
> > /proc/kallsyms).
> >=20
> > Changes from v2 to v3:
> >=20
> > - Use `union loongarch_instruction` instead of explicit bit shifts
> > =C2=A0 applying the relocation.=C2=A0 Suggested by Youling.
> > - For R_LARCH_B26, move the alignment check before the range check to b=
e
> > =C2=A0 consistent with stack pop relocations.=C2=A0 Suggested by Youlin=
g.
> > - Reword the commit message of the 3rd patch.=C2=A0 Suggested by Huacai=
.
>=20
> The log of changes from v1 to v2 should be preserved.
>=20
Retroactively, append the 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).

--=20
Xi Ruoyao <xry111@xry111.site>
School of Aerospace Science and Technology, Xidian University