Message-ID: <d31209ec-ba33-3326-be58-d227c2be8c6d@intel.com>
Date:   Fri, 10 Feb 2023 11:19:56 -0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.6.1
Subject: Re: [PATCH] crypto: x86/aria-avx - fix using avx2 instructions
Content-Language: en-US
To:     Taehee Yoo <ap420073@gmail.com>, linux-crypto@vger.kernel.org,
        herbert@gondor.apana.org.au, davem@davemloft.net, x86@kernel.org
Cc:     erhard_f@mailbox.org
References: <20230210181541.2895144-1-ap420073@gmail.com>
From:   Dave Hansen <dave.hansen@intel.com>
In-Reply-To: <20230210181541.2895144-1-ap420073@gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 2/10/23 10:15, Taehee Yoo wrote:
> vpbroadcastb and vpbroadcastd are not AVX instructions.
> But the aria-avx assembly code contains these instructions.
> So, kernel panic will occur if the aria-avx works on AVX2 unsupported
> CPU.
...
> My CPU supports AVX2.
> So, I disabled AVX2 with QEMU.
> In the VM, lscpu doesn't show AVX2, but kernel panic didn't occur.
> Therefore, I couldn't reproduce kernel panic.
> I will really appreciate it if someone test this patch.

So, someone reported this issue and you _think_ you know what went
wrong.  But, you can't reproduce the issue so it sounds like you're not
confident if this is the right fix or if you are fixing the right
problem in the first place.

We can certainly apply obvious fixes, but it would be *really* nice if
you could try a bit harder to reproduce this.