Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210307051433.2756645-1-maskray@google.com> <20210307062336.GD29191@gate.crashing.org>
In-Reply-To: <20210307062336.GD29191@gate.crashing.org>
From:   =?UTF-8?B?RsSBbmctcnXDrCBTw7JuZw==?= <maskray@google.com>
Date:   Sat, 6 Mar 2021 22:31:15 -0800
Message-ID: <CAFP8O3L91xzZtTQ3gDzpLE=-FvgM3ot5oaBK6jGUZbeT7Pg-RA@mail.gmail.com>
Subject: Re: [PATCH] Replace __toc_start + 0x8000 with .TOC.
To:     Segher Boessenkool <segher@kernel.crashing.org>
Cc:     Michael Ellerman <mpe@ellerman.id.au>,
        linuxppc-dev <linuxppc-dev@lists.ozlabs.org>,
        LKML <linux-kernel@vger.kernel.org>,
        clang-built-linux <clang-built-linux@googlegroups.com>,
        Alexey Kardashevskiy <aik@ozlabs.ru>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Sat, Mar 6, 2021 at 10:25 PM Segher Boessenkool
<segher@kernel.crashing.org> wrote:
>
> Hi!
>
> On Sat, Mar 06, 2021 at 09:14:33PM -0800, Fangrui Song wrote:
> > TOC relocations are like GOT relocations on other architectures.
> > However, unlike other architectures, GNU ld's ppc64 port defines .TOC.
> > relative to the .got output section instead of the linker synthesized
> > .got input section. LLD defines .TOC. as the .got input section plus
> > 0x8000. When CONFIG_PPC_OF_BOOT_TRAMPOLINE=y,
> > arch/powerpc/kernel/prom_init.o is built, and LLD computed .TOC. can be
> > different from __toc_start defined by the linker script.
> >
> > Simplify kernel_toc_addr with asm label .TOC. so that we can get rid of
> > __toc_start.
> >
> > With this change, powernv_defconfig with CONFIG_PPC_OF_BOOT_TRAMPOLINE=y
> > is bootable with LLD. There is still an untriaged issue with Alexey's
> > configuration.
>
> Do you have any explanation why this *does* work, while the original
> doesn't?  Some explanation that says *what* is wrong.  To me it doesn't
> look like the kernel script is.
>
>
> Segher

The kernel code probably wants to access .TOC. (the TOC base symbol)
via __toc_start+0x8000.
If the kernel understood TOC base is different from the linker
understood TOC base (.TOC.), there should be a problem.
By using .TOC. in the kernel code, the two concepts are guaranteed to match.