linux-kernel - Re: [PATCH v3 05/12] arm: vdso: do calculations outside reader loops

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]

Date:   Mon, 30 Oct 2017 14:15:17 +0000
From:   Mark Rutland <mark.rutland@....com>
To:     Mark Salyzyn <salyzyn@...roid.com>
Cc:     linux-kernel@...r.kernel.org, James Morse <james.morse@....com>,
        Russell King <linux@...linux.org.uk>,
        Catalin Marinas <catalin.marinas@....com>,
        Will Deacon <will.deacon@....com>,
        Andy Lutomirski <luto@...capital.net>,
        Dmitry Safonov <dsafonov@...tuozzo.com>,
        John Stultz <john.stultz@...aro.org>,
        Laura Abbott <labbott@...hat.com>,
        Kees Cook <keescook@...omium.org>,
        Ard Biesheuvel <ard.biesheuvel@...aro.org>,
        Andy Gross <andy.gross@...aro.org>,
        Kevin Brodsky <kevin.brodsky@....com>,
        Andrew Pinski <apinski@...ium.com>,
        linux-arm-kernel@...ts.infradead.org,
        Mark Salyzyn <salyzyn@...gle.com>
Subject: Re: [PATCH v3 05/12] arm: vdso: do calculations outside reader loops

On Fri, Oct 27, 2017 at 03:25:40PM -0700, Mark Salyzyn wrote:
> In variable timer reading loops, pick up just the values until all
> are synchronized, then outside of loop pick up cntvct and perform
> calculations to determine final offset, shifted and multiplied
> output value.

So this is all about moving the computation out of the loop, which
sounds sensible...

> +#ifdef ARCH_CLOCK_FIXED_MASK
> +	static const u64 mask = ARCH_CLOCK_FIXED_MASK;
> +#else
> +	u64 mask;
> +#endif

... but this wasn't mentioned, and ARCH_CLOCK_FIXED_MASK doesn't seem to
be defined as of this patch.

It looks like an orthogonal change that should be introduced in a
separate patch.

> +
> +	typeof(((struct vdso_data *)vd)->xtime_clock_sec) sec;

Why do we need to do this typeof() magic?

Can't we settle on a consistent type across arches, or have a typedef in
a header?

Thanks,
Mark.