linux-kernel - Re: [PATCH] crypto: x86/aes-ni: fix AVX detection

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: <aa044725031407b86e7b6edf8a9426166242b8d4.camel@redhat.com>
Date:   Wed, 03 Nov 2021 14:49:48 +0200
From:   Maxim Levitsky <mlevitsk@...hat.com>
To:     linux-kernel@...r.kernel.org
Cc:     "maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)" <x86@...nel.org>,
        "open list:CRYPTO API" <linux-crypto@...r.kernel.org>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Herbert Xu <herbert@...dor.apana.org.au>,
        Borislav Petkov <bp@...en8.de>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Ingo Molnar <mingo@...hat.com>,
        "David S. Miller" <davem@...emloft.net>,
        "H. Peter Anvin" <hpa@...or.com>,
        Tim Chen <tim.c.chen@...ux.intel.com>
Subject: Re: [PATCH] crypto: x86/aes-ni: fix AVX detection

On Wed, 2021-11-03 at 14:46 +0200, Maxim Levitsky wrote:
> Fix two semi-theoretical issues that are present.
> 
> 1. AVX is assumed to be present when AVX2 is present.
>  That can be false in a VM.
>  This can be considered a hypervisor bug,
>  but the kernel should not crash in this case if this is possible.
> 
> 2. YMM state can be soft disabled in XCR0.
> 
> Fix both issues by using 'cpu_has_xfeatures(XFEATURE_MASK_YMM')
> to check for usable AVX support.
> 
> Fixes: d764593af9249 ("crypto: aesni - AVX and AVX2 version of AESNI-GCM encode and decode")
> 
> Signed-off-by: Maxim Levitsky <mlevitsk@...hat.com>

I forgot to mention that Paolo Bonzini helped me with this patch,
especially with the way to detect XCR0 bits.

Best regards,
	Maxim Levitsky

> ---
>  arch/x86/crypto/aesni-intel_glue.c | 25 +++++++++++++------------
>  1 file changed, 13 insertions(+), 12 deletions(-)
> 
> diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c
> index 0fc961bef299c..20db1e500ef6f 100644
> --- a/arch/x86/crypto/aesni-intel_glue.c
> +++ b/arch/x86/crypto/aesni-intel_glue.c
> @@ -1147,24 +1147,25 @@ static int __init aesni_init(void)
>  	if (!x86_match_cpu(aesni_cpu_id))
>  		return -ENODEV;
>  #ifdef CONFIG_X86_64
> -	if (boot_cpu_has(X86_FEATURE_AVX2)) {
> -		pr_info("AVX2 version of gcm_enc/dec engaged.\n");
> -		static_branch_enable(&gcm_use_avx);
> -		static_branch_enable(&gcm_use_avx2);
> -	} else
> -	if (boot_cpu_has(X86_FEATURE_AVX)) {
> -		pr_info("AVX version of gcm_enc/dec engaged.\n");
> +	if (cpu_has_xfeatures(XFEATURE_MASK_YMM, NULL)) {
> +
>  		static_branch_enable(&gcm_use_avx);
> -	} else {
> -		pr_info("SSE version of gcm_enc/dec engaged.\n");
> -	}
> -	if (boot_cpu_has(X86_FEATURE_AVX)) {
> +
> +		if (boot_cpu_has(X86_FEATURE_AVX2)) {
> +			static_branch_enable(&gcm_use_avx2);
> +			pr_info("AVX2 version of gcm_enc/dec engaged.\n");
> +		} else {
> +			pr_info("AVX version of gcm_enc/dec engaged.\n");
> +		}
> +
>  		/* optimize performance of ctr mode encryption transform */
>  		static_call_update(aesni_ctr_enc_tfm, aesni_ctr_enc_avx_tfm);
>  		pr_info("AES CTR mode by8 optimization enabled\n");
> +
> +	} else {
> +		pr_info("SSE version of gcm_enc/dec engaged.\n");
>  	}
>  #endif
> -
>  	err = crypto_register_alg(&aesni_cipher_alg);
>  	if (err)
>  		return err;