linux-kernel - Re: [PATCH 04/22] x86/kvm: Don't call kvm_spurious

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]

Date:   Mon, 15 Jul 2019 20:16:43 +0200
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     Josh Poimboeuf <jpoimboe@...hat.com>
Cc:     x86@...nel.org, linux-kernel@...r.kernel.org,
        Peter Zijlstra <peterz@...radead.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Nick Desaulniers <ndesaulniers@...gle.com>,
        Arnd Bergmann <arnd@...db.de>, Jann Horn <jannh@...gle.com>,
        Randy Dunlap <rdunlap@...radead.org>,
        Radim Krčmář <rkrcmar@...hat.com>
Subject: Re: [PATCH 04/22] x86/kvm: Don't call kvm_spurious_fault() from
 .fixup

On 15/07/19 15:25, Josh Poimboeuf wrote:
> On Mon, Jul 15, 2019 at 03:05:33PM +0200, Paolo Bonzini wrote:
>> On 15/07/19 14:40, Josh Poimboeuf wrote:
>>>>>   * Hardware virtualization extension instructions may fault if a
>>>>>   * reboot turns off virtualization while processes are running.
>>>>> - * Trap the fault and ignore the instruction if that happens.
>>>>> + * If that happens, trap the fault and panic (unless we're rebooting).
>>>> Not sure the comment is better than before, but apar from that
>>> The previous comment didn't seem to match the code, since we only ignore
>>> the instruction if we're rebooting.
>>>
>>
>> "If that happens" refers to "a reboot turns off virtualization while
>> processes are running".
> 
> Ah, makes sense now.  I was reading "if that happens" to mean the fault.
> 
>>  * Usually after catching the fault we just panic; during reboot
>>  * instead the instruction is ignored.
> 
> Yes, that's much clearer.  Assuming you meant to replace the entire
> comment.

No, I didn't. :)  I meant only the last line (otherwise it removes
information on why the fault may happen and the simplest choice is to
ignore).  Thanks for taking care of this!

Paolo


  I also moved it to directly above the macro it's describing:
> 
> 
> asmlinkage void __noreturn kvm_spurious_fault(void);
> 
> /*
>  * Usually after catching the fault we just panic; during reboot
>  * instead the instruction is ignored.
>  */
> #define ____kvm_handle_fault_on_reboot(insn, cleanup_insn)		\
> 
>