Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37;
Message-ID: <4f232dfb-a112-4271-8f43-f85a02c4abeb@suse.com>
Date:   Fri, 8 Dec 2023 20:42:29 +0200
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH -v2] x86/ia32: State that IA32 emulation is disabled
Content-Language: en-US
To:     Borislav Petkov <bp@alien8.de>, X86 ML <x86@kernel.org>
Cc:     LKML <linux-kernel@vger.kernel.org>
References: <20231130155213.1407-1-bp@alien8.de>
 <20231130160903.GJZWizn+dPaaViFVKN@fat_crate.local>
 <20231206172022.GCZXCtVoZtt4t2TLpe@fat_crate.local>
From:   Nikolay Borisov <nik.borisov@suse.com>
In-Reply-To: <20231206172022.GCZXCtVoZtt4t2TLpe@fat_crate.local>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk


On 6.12.23 г. 19:20 ч., Borislav Petkov wrote:
> From: "Borislav Petkov (AMD)" <bp@alien8.de>
> Date: Thu, 30 Nov 2023 16:39:33 +0100
> 
> Issue a short message once, on the first try to load a 32-bit process to
> save people time when wondering why it won't load and trying to execute
> it, would say:
> 
>    -bash: ./strsep32: cannot execute binary file: Exec format error
> 
> Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
> Cc: Nikolay Borisov <nik.borisov@suse.com>

LGTM:

Reviewed-by: Nikolay Borisov <nik.borisov@suse.com>


> ---
>   arch/x86/include/asm/elf.h  |  2 +-
>   arch/x86/include/asm/ia32.h | 11 ++++++++++-
>   2 files changed, 11 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/x86/include/asm/elf.h b/arch/x86/include/asm/elf.h
> index a0234dfd1031..1e16bd5ac781 100644
> --- a/arch/x86/include/asm/elf.h
> +++ b/arch/x86/include/asm/elf.h
> @@ -150,7 +150,7 @@ do {						\
>   	((x)->e_machine == EM_X86_64)
>   
>   #define compat_elf_check_arch(x)					\
> -	((elf_check_arch_ia32(x) && ia32_enabled()) ||			\
> +	((elf_check_arch_ia32(x) && ia32_enabled_verbose()) ||		\
>   	 (IS_ENABLED(CONFIG_X86_X32_ABI) && (x)->e_machine == EM_X86_64))
>   
>   static inline void elf_common_init(struct thread_struct *t,
> diff --git a/arch/x86/include/asm/ia32.h b/arch/x86/include/asm/ia32.h
> index 5a2ae24b1204..094886a8717e 100644
> --- a/arch/x86/include/asm/ia32.h
> +++ b/arch/x86/include/asm/ia32.h
> @@ -2,7 +2,6 @@
>   #ifndef _ASM_X86_IA32_H
>   #define _ASM_X86_IA32_H
>   
> -
>   #ifdef CONFIG_IA32_EMULATION
>   
>   #include <linux/compat.h>
> @@ -84,4 +83,14 @@ static inline bool ia32_enabled(void)
>   
>   #endif
>   
> +static inline bool ia32_enabled_verbose(void)
> +{
> +	bool enabled = ia32_enabled();
> +
> +	if (IS_ENABLED(CONFIG_IA32_EMULATION) && !enabled)

nit: But why check config_ia32_emulation explicitly, its state should 
already be accounted for in ia32_enabled's value? So this check can 
simply be 'if (!enabled)'

> +		pr_notice_once("32-bit emulation disabled. You can reenable with ia32_emulation=on\n");
> +
> +	return enabled;
> +}
> +
>   #endif /* _ASM_X86_IA32_H */