linux-kernel - Re: [PATCH v1 0/3] x86/boot: Supply boot_param in rdi instead of rsi from startup

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: <9FE74F68-5CF9-4EFE-8271-551ECC9087B9@zytor.com>
Date: Mon, 23 Jun 2025 17:43:27 -0700
From: "H. Peter Anvin" <hpa@...or.com>
To: Brian Gerst <brgerst@...il.com>, Khalid Ali <khaliidcaliy@...il.com>
CC: tglx@...utronix.de, mingo@...hat.com, bp@...en8.de,
        dave.hansen@...ux.intel.com, ubizjak@...il.com, x86@...nel.org,
        linux-efi@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1 0/3] x86/boot: Supply boot_param in rdi instead of rsi from startup_64()

On June 23, 2025 12:24:50 PM PDT, Brian Gerst <brgerst@...il.com> wrote:
>On Mon, Jun 23, 2025 at 2:40 PM Khalid Ali <khaliidcaliy@...il.com> wrote:
>>
>> > This is also invoked by some external bootloaders that boot the ELF
>> > image directly, even though this is strongly discouraged.
>> >
>> > Therefore this patchset is NAKed with extreme prejudice.
>>
>> Thanks both of you peter and brian,
>>
>> however, the boot protocol document saying "%rsi must hold the base address of the struct boot_params",
>> it doesn't mention why. Maybe the document needs update to justify the reasons. I wouldn't have known it
>> if you didn't tell me, so this shouldn't confuse anyone else.
>
>The use of RSI was inherited from the 32-bit kernel, but the real
>reason is lost to history.  It's just always been that way and there
>is no compelling reason to change it.
>
>
>Brian Gerst
>

The reason isn't lost to history: I picked %esi because I found that none of the weird bootloaders which looked the protected mode jump clobbered that particular register.