linux-kernel - Re: [PATCH v2 18/18] perf: arm_spe: Disallow userspace profiling when arm_kernel_unmapped_at

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: <20171201164943.GJ14823@arm.com>
Date:   Fri, 1 Dec 2017 16:49:44 +0000
From:   Will Deacon <will.deacon@....com>
To:     Mark Rutland <mark.rutland@....com>
Cc:     linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        catalin.marinas@....com, ard.biesheuvel@...aro.org,
        sboyd@...eaurora.org, dave.hansen@...ux.intel.com,
        keescook@...omium.org, msalter@...hat.com, labbott@...hat.com,
        tglx@...utronix.de
Subject: Re: [PATCH v2 18/18] perf: arm_spe: Disallow userspace profiling
 when arm_kernel_unmapped_at_el0()

On Fri, Dec 01, 2017 at 12:15:06PM +0000, Mark Rutland wrote:
> On Thu, Nov 30, 2017 at 04:39:46PM +0000, Will Deacon wrote:
> > When running with the kernel unmapped whilst at EL0, the virtually-addressed
> > SPE buffer is also unmapped, which can lead to buffer faults if userspace
> > profiling is enabled.
> > 
> > This patch prohibits SPE profiling of userspace when
> > arm_kernel_unmapped_at_el0().
> > 
> > Signed-off-by: Will Deacon <will.deacon@....com>
> > ---
> >  drivers/perf/arm_spe_pmu.c | 7 +++++++
> >  1 file changed, 7 insertions(+)
> > 
> > diff --git a/drivers/perf/arm_spe_pmu.c b/drivers/perf/arm_spe_pmu.c
> > index 8ce262fc2561..c028db8973a4 100644
> > --- a/drivers/perf/arm_spe_pmu.c
> > +++ b/drivers/perf/arm_spe_pmu.c
> > @@ -675,6 +675,13 @@ static int arm_spe_pmu_event_init(struct perf_event *event)
> >  		return -EOPNOTSUPP;
> >  
> >  	/*
> > +	 * If kernelspace is unmapped when running at EL0, then the SPE
> > +	 * buffer will fault and prematurely terminate the AUX session.
> > +	 */
> > +	if (arm64_kernel_unmapped_at_el0() && !attr->exclude_user)
> > +		dev_warn_once(&spe_pmu->pdev->dev, "unable to write to profiling buffer from EL0. Try passing \"kaiser=off\" on the kernel command line");
> 
> The commit messages sats this prohibits profiling, but we simply log a
> message.
> 
> I take it you meant to return an error code, too?

To be honest with you, I've been changing my mind a lot about what to do
here and the code has ended up being a bit of a mess after I've butchered
it repeatedly.

The fact remains that there aren't any SPE-capable CPUs shipping at the
moment, so I'm inclined just to fail the probe for now and we can look
at whether or not we can do better when we've got some hardware to play
with.

And I'll add the missing newline.

Thanks,

Will