linux-kernel - Re: [PATCH 1/3] lib/vdso: Delay mask application in do

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:   Tue, 25 Jun 2019 19:02:13 +0200 (CEST)
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Vincenzo Frascino <vincenzo.frascino@....com>
cc:     linux-arch@...r.kernel.org,
        LAK <linux-arm-kernel@...ts.infradead.org>,
        LKML <linux-kernel@...r.kernel.org>, linux-mips@...r.kernel.org,
        linux-kselftest@...r.kernel.org, catalin.marinas@....com,
        Will Deacon <will.deacon@....com>,
        Arnd Bergmann <arnd@...db.de>, linux@...linux.org.uk,
        Ralf Baechle <ralf@...ux-mips.org>, paul.burton@...s.com,
        Daniel Lezcano <daniel.lezcano@...aro.org>,
        salyzyn@...roid.com, pcc@...gle.com, shuah@...nel.org,
        0x7f454c46@...il.com, linux@...musvillemoes.dk,
        huw@...eweavers.com, sthotton@...vell.com, andre.przywara@....com,
        Andy Lutomirski <luto@...nel.org>
Subject: Re: [PATCH 1/3] lib/vdso: Delay mask application in do_hres()

On Tue, 25 Jun 2019, Vincenzo Frascino wrote:

CC+ Andy

> do_hres() in the vDSO generic library masks the hw counter value
> immediately after reading it.
> 
> Postpone the mask application after checking if the syscall fallback is
> enabled, in order to be able to detect a possible fallback for the
> architectures that have masks smaller than ULLONG_MAX.

Right. This only worked on x86 because the mask is there ULLONG_MAX for all
VDSO capable clocksources, i.e. that ever worked just by chance.

As we talked about that already yesterday, I tested this on a couple of
machines and as expected the outcome is uarch dependent. Minimal deviations
to both sides and some machines do not show any change at all. I doubt it's
possible to come up with a solution which makes all uarchs go faster
magically.

Though, thinking about it, we could remove the mask operation completely on
X86. /me runs tests

Thanks,

	tglx


> Signed-off-by: Vincenzo Frascino <vincenzo.frascino@....com>
> ---
>  lib/vdso/gettimeofday.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/lib/vdso/gettimeofday.c b/lib/vdso/gettimeofday.c
> index ef28cc5d7bff..ee1221ba1d32 100644
> --- a/lib/vdso/gettimeofday.c
> +++ b/lib/vdso/gettimeofday.c
> @@ -35,12 +35,12 @@ static int do_hres(const struct vdso_data *vd, clockid_t clk,
>  
>  	do {
>  		seq = vdso_read_begin(vd);
> -		cycles = __arch_get_hw_counter(vd->clock_mode) &
> -			vd->mask;
> +		cycles = __arch_get_hw_counter(vd->clock_mode);
>  		ns = vdso_ts->nsec;
>  		last = vd->cycle_last;
>  		if (unlikely((s64)cycles < 0))
>  			return clock_gettime_fallback(clk, ts);
> +		cycles &= vd->mask;
>  		if (cycles > last)
>  			ns += (cycles - last) * vd->mult;
>  		ns >>= vd->shift;
> -- 
> 2.22.0
> 
>