linux-kernel - Re: [PATCH v4 03/12] x86/mtrr: support setting MTRR state for software defined MTRRs

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: <0552d5a7-b6b7-e3e0-4d20-a64d455c2948@suse.com>
Date:   Tue, 21 Mar 2023 16:49:36 +0100
From:   Juergen Gross <jgross@...e.com>
To:     Borislav Petkov <bp@...en8.de>
Cc:     linux-kernel@...r.kernel.org, x86@...nel.org,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        "H. Peter Anvin" <hpa@...or.com>
Subject: Re: [PATCH v4 03/12] x86/mtrr: support setting MTRR state for
 software defined MTRRs

On 21.03.23 11:30, Borislav Petkov wrote:
> On Tue, Mar 21, 2023 at 07:00:58AM +0100, Juergen Gross wrote:
>> I guess you are asking because the next test seems to catch the same case?
>>
>> I think it doesn't, e.g. for the case of unknown hypervisors (which shows that
>> X86_HYPER_NATIVE in theory should be named X86_HYPER_NATIVE_OR_UNKNOWN, or it
>> should be split into X86_HYPER_NATIVE and X86_HYPER_UNKNOWN).
> 
> Yeah, we don't care about unknown hypervisors. They'll crash'n'burn
> anyway.

Okay, I'll drop that test.

> My intent is to have every case properly documented with a comment above it
> instead of one huge compound conditional.
> 
>> It basically doesn't matter.
> 
> It doesn't matter now. Until someone decides to "redefine" how MTRRs
> should be done again because the next representative from the virt zoo
> decided to do magic pink ponies.
> 
> I'm not taking any chances anymore judging by the amount of crap that
> gets sent into arch/x86/ to support some weird guest contraption.
> 
>> The only possibility of mtrr_state.enabled to be set at this point is a
>> previous call of mtrr_overwrite_state().
> 
> Sure, pls make it explicit and defensive so that it is perfectly clear
> what's going on.

Okay, will do the modification you were suggesting.


Juergen

Download attachment "OpenPGP_0xB0DE9DD628BF132F.asc" of type "application/pgp-keys" (3099 bytes)

Download attachment "OpenPGP_signature" of type "application/pgp-signature" (496 bytes)