linux-kernel - Re: [PATCH v5 2/4] riscv: Introduce CONFIG

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <43ef1623-2186-aa07-cfa5-d618aa4f09c0@ghiti.fr>
Date:   Thu, 11 Jun 2020 15:43:56 -0400
From:   Alex Ghiti <alex@...ti.fr>
To:     Jerome Forissier <jerome@...issier.org>,
        Michael Ellerman <mpe@...erman.id.au>,
        Benjamin Herrenschmidt <benh@...nel.crashing.org>,
        Paul Mackerras <paulus@...ba.org>,
        Paul Walmsley <paul.walmsley@...ive.com>,
        Palmer Dabbelt <palmer@...belt.com>,
        Albert Ou <aou@...s.berkeley.edu>,
        Anup Patel <Anup.Patel@....com>,
        Atish Patra <Atish.Patra@....com>,
        Zong Li <zong.li@...ive.com>, linux-kernel@...r.kernel.org,
        linuxppc-dev@...ts.ozlabs.org, linux-riscv@...ts.infradead.org
Cc:     Anup Patel <anup@...infault.org>
Subject: Re: [PATCH v5 2/4] riscv: Introduce CONFIG_RELOCATABLE

Hi Jerome,

Le 6/10/20 à 10:10 AM, Jerome Forissier a écrit :
> On 6/7/20 9:59 AM, Alexandre Ghiti wrote:
> [...]
>
>> +config RELOCATABLE
>> +	bool
>> +	depends on MMU
>> +	help
>> +          This builds a kernel as a Position Independent Executable (PIE),
>> +          which retains all relocation metadata required to relocate the
>> +          kernel binary at runtime to a different virtual address than the
>> +          address it was linked at.
>> +          Since RISCV uses the RELA relocation format, this requires a
>> +          relocation pass at runtime even if the kernel is loaded at the
>> +          same address it was linked at.
> Is this true? I thought that the GNU linker would write the "proper"
> values by default, contrary to the LLVM linker (ld.lld) which would need
> a special flag: --apply-dynamic-relocs (by default the relocated places
> are set to zero). At least, it is my experience with Aarch64 on a
> different project. So, sorry if I'm talking nonsense here -- I have not
> looked at the details.
>
>

It seems that you're right, at least for aarch64 since they specifically 
specify the --no-apply-dynamic-relocs option. I retried to boot without 
relocating at runtime, and it fails on riscv. Can this be arch specific ?

Thanks,

Alex