Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
Date:   Thu, 20 Oct 2022 16:30:26 +0300
From:   Evgeniy Baskov <baskov@ispras.ru>
To:     Andrew Cooper <Andrew.Cooper3@citrix.com>
Cc:     Ard Biesheuvel <ardb@kernel.org>, Borislav Petkov <bp@alien8.de>,
        Andy Lutomirski <luto@kernel.org>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Alexey Khoroshilov <khoroshilov@ispras.ru>,
        lvc-project@linuxtesting.org, x86@kernel.org,
        linux-efi@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-hardening@vger.kernel.org
Subject: Re: [PATCH 06/16] x86/boot: Setup memory protection for bzImage code
In-Reply-To: <58b9f97c-36d5-6b9e-5336-14ad3a09ad18@citrix.com>
References: <cover.1662459668.git.baskov@ispras.ru>
 <2fd61a79a1e6885dc47ec826b62a936dd88a0a16.1662459668.git.baskov@ispras.ru>
 <58b9f97c-36d5-6b9e-5336-14ad3a09ad18@citrix.com>
User-Agent: Roundcube Webmail/1.4.4
Message-ID: <ed8c18e42870da96afc9ce1222816189@ispras.ru>
Content-Type: text/plain; charset=US-ASCII;
 format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 2022-10-19 10:57, Andrew Cooper wrote:
> On 06/09/2022 11:41, Evgeniy Baskov wrote:
>> diff --git a/arch/x86/boot/compressed/head_64.S 
>> b/arch/x86/boot/compressed/head_64.S
>> index 5273367283b7..889ca7176aa7 100644
>> --- a/arch/x86/boot/compressed/head_64.S
>> +++ b/arch/x86/boot/compressed/head_64.S
>> @@ -602,6 +603,28 @@ SYM_FUNC_START_LOCAL_NOALIGN(.Lrelocated)
>>  	jmp	*%rax
>>  SYM_FUNC_END(.Lrelocated)
>> 
>> +SYM_FUNC_START_LOCAL_NOALIGN(startup32_enable_nx_if_supported)
>> +	pushq	%rbx
>> +
>> +	leaq	has_nx(%rip), %rcx
>> +
>> +	mov	$0x80000001, %eax
>> +	cpuid
>> +	btl	$20, %edx
> 
> btl $(X86_FEATURE_NX & 31), %edx
> 
> But also need to check for the availability of the extended leaf in the
> first place.

Yes, thank you for suggestion, that looks more readable. I will
also add the leaf node check. Is there any processor though that
supports long mode and does not support 0x80000001 leaf node?

> 
>> +	jnc	.Lnonx
>> +
>> +	movl	$1, (%rcx)
> 
> Your pointer has been clobbered with some feature flags.

Thanks, I apparently forgot to include fix for this into a patch set...

> 
> movl $1, has_nx(%rip)
> 
> will work fine without needing the intermediary lea.
> 
> ~Andrew