linux-kernel - Re: [PATCH v5] x86/setup: Allow passing RNG seeds via e820 setup table

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: <709B69E7-DEC8-428C-9555-254E32C834A9@zytor.com>
Date:   Fri, 01 Jul 2022 11:52:03 -0700
From:   "H. Peter Anvin" <hpa@...or.com>
To:     "Jason A. Donenfeld" <Jason@...c4.com>
CC:     tglx@...utronix.de, mingo@...hat.com, bp@...en8.de,
        dave.hansen@...ux.intel.com, x86@...nel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v5] x86/setup: Allow passing RNG seeds via e820 setup table

On July 1, 2022 11:42:03 AM PDT, "Jason A. Donenfeld" <Jason@...c4.com> wrote:
>Hi Peter,
>
>On Fri, Jul 01, 2022 at 11:25:49AM -0700, H. Peter Anvin wrote:
>> Please correct the incredibly confusing title of this patch.
>> 
>> The setup_data linked list has nothing to do with memory types (e820),
>> except that memory types is one kind of data that can be passed on by
>> this mechanism. This title makes it sound like you are passing random
>> data in as a memory type, which could make some bizarre sense if it
>> were, say, some kind of ring buffer preconfigured by the BIOS/VMM/boot
>> loader to be continually overwritten with new random data.
>
>Yea that is pretty confusing. Alright, I'll retitle that to something
>like:
>
>    x86/setup: Allow passing RNG seeds via setup data
>
>Hopefully that matches more, considering the function that does the work
>is called "parse_setup_data".
>
>(I'll wait some more time before posting yet another v+1 with that
>change, in case others have more feedback.)
>
>Jason

We usually refer to it as setup_data (with the underscore) to indicate that it is a name and not a description.