linux-kernel - Re: [RFC PATCH] arm64/fpsimd: Don't disable softirq when touching FPSIMD/SVE state

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <10464811-e050-5ea1-6763-4362cb1c7c7c@arm.com>
Date:   Mon, 18 Feb 2019 16:32:14 +0000
From:   Julien Grall <julien.grall@....com>
To:     linux-arm-kernel@...ts.infradead.org
Cc:     bigeasy@...utronix.de, Dave.Martin@....com,
        linux-rt-users@...r.kernel.org, catalin.marinas@....com,
        will.deacon@....com, ard.biesheuvel@...aro.org,
        linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH] arm64/fpsimd: Don't disable softirq when touching
 FPSIMD/SVE state

Hi all,

On 08/02/2019 16:55, Julien Grall wrote:
> @@ -1094,9 +1113,7 @@ void kernel_neon_begin(void)
>   	/* Invalidate any task state remaining in the fpsimd regs: */
>   	fpsimd_flush_cpu_state();
>   
> -	preempt_disable();
> -
> -	local_bh_enable();
> +	kernel_neon_enable();

I found one error in the patch. We should not re-enable NEON from 
kernel_neon_begin(). I will resend a patch once the thread settle down.

Cheers,

-- 
Julien Grall