All of lore.kernel.org
 help / color / mirror / Atom feed
From: Will Deacon <will.deacon@arm.com>
To: Catalin Marinas <catalin.marinas@arm.com>
Cc: Kristina Martsenko <kristina.martsenko@arm.com>,
	Ramana Radhakrishnan <ramana.radhakrishnan@foss.arm.com>,
	"linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>,
	Mark Rutland <Mark.Rutland@arm.com>,
	"linux-arch@vger.kernel.org" <linux-arch@vger.kernel.org>,
	Andrew Jones <drjones@redhat.com>,
	Jacob Bramley <Jacob.Bramley@arm.com>,
	Arnd Bergmann <arnd@arndb.de>,
	Ard Biesheuvel <ard.biesheuvel@linaro.org>,
	Marc Zyngier <Marc.Zyngier@arm.com>,
	Adam Wallis <awallis@codeaurora.org>,
	Suzuki Poulose <Suzuki.Poulose@arm.com>,
	Christoffer Dall <Christoffer.Dall@arm.com>,
	"kvmarm@lists.cs.columbia.edu" <kvmarm@lists.cs.columbia.edu>,
	Amit Kachhap <Amit.Kachhap@arm.com>,
	Dave P Martin <Dave.Martin@arm.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v5 11/17] arm64: docs: document pointer authentication
Date: Fri, 19 Oct 2018 13:22:05 +0100	[thread overview]
Message-ID: <20181019122204.GE14246@arm.com> (raw)
In-Reply-To: <20181019113556.ljbdmjo5pdw7muvz@mbp>

On Fri, Oct 19, 2018 at 12:35:56PM +0100, Catalin Marinas wrote:
> On Tue, Oct 16, 2018 at 05:14:39PM +0100, Kristina Martsenko wrote:
> > On 05/10/2018 10:04, Ramana Radhakrishnan wrote:
> > > On 05/10/2018 09:47, Kristina Martsenko wrote:
> > The other special case is the XPACLRI instruction, which is also in the
> > HINT space. Currently it will trap and KVM will inject an exception into
> > the guest. We should probably change this to NOP instead, as that's what
> > applications will expect. Unfortunately there is no EnIA-like control to
> > make it NOP.
> 
> Very good catch. Basically if EL2 doesn't know about ptr auth (older
> distro), EL1 may or may not know but leaves SCTLR_EL1 disabled (based on
> CPUID), the default HCR_EL2 is to trap (I'm ignoring EL3 as that's like
> to have ptr auth enabled, being built for the specific HW). So a user
> app considering XPACLRI a NOP (or inoffensive) will get a SIGILL
> (injected by the guest kernel following the injection of "Unknown
> reason" exception by KVM).
> 
> Ramana, is XPACLRI commonly generated by gcc and expects it to be a NOP?
> Could we restrict it to only being used at run-time if the corresponding
> HWCAP is set? This means redefining this instruction as no longer in the
> NOP space.

My main worry is that this instruction is used when unwinding C++
exceptions, so I think we'll see it fairly often.

Effectively, the architecture means these instructions can result in a
SIGILL if they are used under an OS/hypervisor that doesn't know about
the feature (i.e. any mainline kernel release so far). I think that's a
massive problem for the current implementation in GCC. Worse, if
distributions are currently shipping binaries built with this, they
basically have a ticking bomb in their applications where things will
start crashing when they encounter CPUs that implement pointer
authentication.

Ramana: do you know whether people are building binaries with this stuff
enabled by default?

Will

WARNING: multiple messages have this Message-ID (diff)
From: Will Deacon <will.deacon@arm.com>
To: Catalin Marinas <catalin.marinas@arm.com>
Cc: Kristina Martsenko <kristina.martsenko@arm.com>,
	Ramana Radhakrishnan <ramana.radhakrishnan@foss.arm.com>,
	"linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>,
	Mark Rutland <Mark.Rutland@arm.com>,
	"linux-arch@vger.kernel.org" <linux-arch@vger.kernel.org>,
	Andrew Jones <drjones@redhat.com>,
	Jacob Bramley <Jacob.Bramley@arm.com>,
	Arnd Bergmann <arnd@arndb.de>,
	Ard Biesheuvel <ard.biesheuvel@linaro.org>,
	Marc Zyngier <Marc.Zyngier@arm.com>,
	Adam Wallis <awallis@codeaurora.org>,
	Suzuki Poulose <Suzuki.Poulose@arm.com>,
	Christoffer Dall <Christoffer.Dall@arm.com>,
	"kvmarm@lists.cs.columbia.edu" <kvmarm@lists.cs.columbia.edu>,
	Amit Kachhap <Amit.Kachhap@arm.com>,
	Dave P Martin <Dave.Martin@arm.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Kees Cook <keescook@chromium.org>
Subject: Re: [PATCH v5 11/17] arm64: docs: document pointer authentication
Date: Fri, 19 Oct 2018 13:22:05 +0100	[thread overview]
Message-ID: <20181019122204.GE14246@arm.com> (raw)
Message-ID: <20181019122205.GGCM5UCezhwnpi3tCH_EmUNxqeb9Nu_KZWpHmX7JF-0@z> (raw)
In-Reply-To: <20181019113556.ljbdmjo5pdw7muvz@mbp>

On Fri, Oct 19, 2018 at 12:35:56PM +0100, Catalin Marinas wrote:
> On Tue, Oct 16, 2018 at 05:14:39PM +0100, Kristina Martsenko wrote:
> > On 05/10/2018 10:04, Ramana Radhakrishnan wrote:
> > > On 05/10/2018 09:47, Kristina Martsenko wrote:
> > The other special case is the XPACLRI instruction, which is also in the
> > HINT space. Currently it will trap and KVM will inject an exception into
> > the guest. We should probably change this to NOP instead, as that's what
> > applications will expect. Unfortunately there is no EnIA-like control to
> > make it NOP.
> 
> Very good catch. Basically if EL2 doesn't know about ptr auth (older
> distro), EL1 may or may not know but leaves SCTLR_EL1 disabled (based on
> CPUID), the default HCR_EL2 is to trap (I'm ignoring EL3 as that's like
> to have ptr auth enabled, being built for the specific HW). So a user
> app considering XPACLRI a NOP (or inoffensive) will get a SIGILL
> (injected by the guest kernel following the injection of "Unknown
> reason" exception by KVM).
> 
> Ramana, is XPACLRI commonly generated by gcc and expects it to be a NOP?
> Could we restrict it to only being used at run-time if the corresponding
> HWCAP is set? This means redefining this instruction as no longer in the
> NOP space.

My main worry is that this instruction is used when unwinding C++
exceptions, so I think we'll see it fairly often.

Effectively, the architecture means these instructions can result in a
SIGILL if they are used under an OS/hypervisor that doesn't know about
the feature (i.e. any mainline kernel release so far). I think that's a
massive problem for the current implementation in GCC. Worse, if
distributions are currently shipping binaries built with this, they
basically have a ticking bomb in their applications where things will
start crashing when they encounter CPUs that implement pointer
authentication.

Ramana: do you know whether people are building binaries with this stuff
enabled by default?

Will

WARNING: multiple messages have this Message-ID (diff)
From: will.deacon@arm.com (Will Deacon)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v5 11/17] arm64: docs: document pointer authentication
Date: Fri, 19 Oct 2018 13:22:05 +0100	[thread overview]
Message-ID: <20181019122204.GE14246@arm.com> (raw)
In-Reply-To: <20181019113556.ljbdmjo5pdw7muvz@mbp>

On Fri, Oct 19, 2018 at 12:35:56PM +0100, Catalin Marinas wrote:
> On Tue, Oct 16, 2018 at 05:14:39PM +0100, Kristina Martsenko wrote:
> > On 05/10/2018 10:04, Ramana Radhakrishnan wrote:
> > > On 05/10/2018 09:47, Kristina Martsenko wrote:
> > The other special case is the XPACLRI instruction, which is also in the
> > HINT space. Currently it will trap and KVM will inject an exception into
> > the guest. We should probably change this to NOP instead, as that's what
> > applications will expect. Unfortunately there is no EnIA-like control to
> > make it NOP.
> 
> Very good catch. Basically if EL2 doesn't know about ptr auth (older
> distro), EL1 may or may not know but leaves SCTLR_EL1 disabled (based on
> CPUID), the default HCR_EL2 is to trap (I'm ignoring EL3 as that's like
> to have ptr auth enabled, being built for the specific HW). So a user
> app considering XPACLRI a NOP (or inoffensive) will get a SIGILL
> (injected by the guest kernel following the injection of "Unknown
> reason" exception by KVM).
> 
> Ramana, is XPACLRI commonly generated by gcc and expects it to be a NOP?
> Could we restrict it to only being used at run-time if the corresponding
> HWCAP is set? This means redefining this instruction as no longer in the
> NOP space.

My main worry is that this instruction is used when unwinding C++
exceptions, so I think we'll see it fairly often.

Effectively, the architecture means these instructions can result in a
SIGILL if they are used under an OS/hypervisor that doesn't know about
the feature (i.e. any mainline kernel release so far). I think that's a
massive problem for the current implementation in GCC. Worse, if
distributions are currently shipping binaries built with this, they
basically have a ticking bomb in their applications where things will
start crashing when they encounter CPUs that implement pointer
authentication.

Ramana: do you know whether people are building binaries with this stuff
enabled by default?

Will

  parent reply	other threads:[~2018-10-19 12:22 UTC|newest]

Thread overview: 149+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-05  8:47 [PATCH 00/17] ARMv8.3 pointer authentication support Kristina Martsenko
2018-10-05  8:47 ` Kristina Martsenko
2018-10-05  8:47 ` [PATCH v5 01/17] arm64: add pointer authentication register bits Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-11 16:28   ` Will Deacon
2018-10-11 16:28     ` Will Deacon
2018-10-12  8:53     ` Mark Rutland
2018-10-12  8:53       ` Mark Rutland
2018-10-12  8:56       ` Will Deacon
2018-10-12  8:56         ` Will Deacon
2018-10-12  9:50         ` Mark Rutland
2018-10-12  9:50           ` Mark Rutland
2018-10-05  8:47 ` [PATCH v5 02/17] arm64/kvm: consistently handle host HCR_EL2 flags Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47 ` [PATCH v5 03/17] arm64/kvm: hide ptrauth from guests Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47 ` [PATCH v5 04/17] arm64: Don't trap host pointer auth use to EL2 Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47 ` [PATCH v5 05/17] arm64/cpufeature: detect pointer authentication Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47 ` [PATCH v5 06/17] asm-generic: mm_hooks: allow hooks to be overridden individually Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47 ` [PATCH v5 07/17] arm64: add basic pointer authentication support Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-11 16:00   ` Suzuki K Poulose
2018-10-11 16:00     ` Suzuki K Poulose
2018-10-19 11:15   ` Catalin Marinas
2018-10-19 11:15     ` Catalin Marinas
2018-10-19 11:15     ` Catalin Marinas
2018-10-19 11:24     ` Will Deacon
2018-10-19 11:24       ` Will Deacon
2018-10-19 15:36       ` Kees Cook
2018-10-19 15:36         ` Kees Cook
2018-10-19 15:36         ` Kees Cook
2018-10-19 15:49         ` Will Deacon
2018-10-19 15:49           ` Will Deacon
2018-10-19 15:49           ` Will Deacon
2018-10-19 16:05           ` Kees Cook
2018-10-19 16:05             ` Kees Cook
2018-10-19 16:05             ` Kees Cook
2018-10-19 16:16             ` Will Deacon
2018-10-19 16:16               ` Will Deacon
2018-10-19 16:16               ` Will Deacon
2018-10-19 15:54         ` Mark Rutland
2018-10-19 15:54           ` Mark Rutland
2018-10-19 15:54           ` Mark Rutland
2018-10-19 16:49       ` Cyrill Gorcunov
2018-10-19 16:49         ` Cyrill Gorcunov
2018-10-19 16:49         ` Cyrill Gorcunov
2018-11-14 18:11       ` Will Deacon
2018-11-14 18:11         ` Will Deacon
2018-11-15 10:25         ` Dave Martin
2018-11-15 10:25           ` Dave Martin
2018-10-23  8:36     ` Ramana Radhakrishnan
2018-10-23  8:36       ` Ramana Radhakrishnan
2018-10-23  8:36       ` Ramana Radhakrishnan
2018-10-23  8:36       ` Ramana Radhakrishnan
2018-10-23 10:20       ` Will Deacon
2018-10-23 10:20         ` Will Deacon
2018-10-23 10:20         ` Will Deacon
2018-10-23 10:20         ` Will Deacon
2018-10-05  8:47 ` [PATCH v5 08/17] arm64: expose user PAC bit positions via ptrace Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47 ` [PATCH v5 09/17] arm64: perf: strip PAC when unwinding userspace Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47 ` [PATCH v5 10/17] arm64: enable pointer authentication Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47 ` [PATCH v5 11/17] arm64: docs: document " Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  9:04   ` Ramana Radhakrishnan
2018-10-05  9:04     ` Ramana Radhakrishnan
2018-10-16 16:14     ` Kristina Martsenko
2018-10-16 16:14       ` Kristina Martsenko
2018-10-19 11:35       ` Catalin Marinas
2018-10-19 11:35         ` Catalin Marinas
2018-10-19 11:35         ` Catalin Marinas
2018-10-19 11:47         ` Marc Zyngier
2018-10-19 11:47           ` Marc Zyngier
2018-10-19 11:47           ` Marc Zyngier
2018-10-19 12:22         ` Will Deacon [this message]
2018-10-19 12:22           ` Will Deacon
2018-10-19 12:22           ` Will Deacon
2018-10-19 14:42         ` Kristina Martsenko
2018-10-19 14:42           ` Kristina Martsenko
2018-10-19 14:42           ` Kristina Martsenko
2018-10-19 15:10           ` Catalin Marinas
2018-10-19 15:10             ` Catalin Marinas
2018-10-19 15:10             ` Catalin Marinas
2018-10-19 17:45             ` Will Deacon
2018-10-19 17:45               ` Will Deacon
2018-10-19 17:45               ` Will Deacon
2018-11-02  6:02               ` Jon Masters
2018-11-02  6:02                 ` Jon Masters
2018-11-02  6:02                 ` Jon Masters
2018-10-24 10:56         ` Ramana Radhakrishnan
2018-10-24 10:56           ` Ramana Radhakrishnan
2018-10-15 22:35   ` Kees Cook
2018-10-15 22:35     ` Kees Cook
2018-11-02  9:46     ` Ramana Radhakrishnan
2018-11-02  9:46       ` Ramana Radhakrishnan
2018-11-02  9:46       ` Ramana Radhakrishnan
2018-10-05  8:47 ` [RFC 12/17] arm64: move ptrauth keys to thread_info Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-19 11:38   ` Catalin Marinas
2018-10-19 11:38     ` Catalin Marinas
2018-10-05  8:47 ` [RFC 13/17] arm64: install user ptrauth keys at kernel exit time Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47 ` [RFC 14/17] arm64: unwind: strip PAC from kernel addresses Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  8:47 ` [RFC 15/17] arm64: enable ptrauth earlier Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-06 12:51   ` Amit Kachhap
2018-10-06 12:51     ` Amit Kachhap
2018-10-05  8:47 ` [RFC 16/17] arm64: initialize and switch ptrauth kernel keys Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-06 12:56   ` Amit Kachhap
2018-10-06 12:56     ` Amit Kachhap
2018-10-05  8:47 ` [RFC 17/17] arm64: compile the kernel with ptrauth -msign-return-address Kristina Martsenko
2018-10-05  8:47   ` Kristina Martsenko
2018-10-05  9:01   ` Ramana Radhakrishnan
2018-10-05  9:01     ` Ramana Radhakrishnan
2018-10-11 14:00     ` Kristina Martsenko
2018-10-11 14:00       ` Kristina Martsenko
2018-10-11 14:23   ` Vladimir Murzin
2018-10-11 14:23     ` Vladimir Murzin
2018-10-15 22:38     ` Kees Cook
2018-10-15 22:38       ` Kees Cook
2018-10-15 22:38       ` Kees Cook
2018-10-15 22:42 ` [PATCH 00/17] ARMv8.3 pointer authentication support Kees Cook
2018-10-15 22:42   ` Kees Cook
2018-11-13 16:17   ` Kristina Martsenko
2018-11-13 16:17     ` Kristina Martsenko
2018-11-13 23:09     ` Kees Cook
2018-11-13 23:09       ` Kees Cook
2018-11-14 15:54       ` Kristina Martsenko
2018-11-14 15:54         ` Kristina Martsenko
2018-11-14 21:47       ` Mark Rutland
2018-11-14 21:47         ` Mark Rutland
2018-11-14 22:48         ` Kees Cook
2018-11-14 22:48           ` Kees Cook
2018-11-14 22:48           ` Kees Cook
2018-10-19 12:36 ` Will Deacon
2018-10-19 12:36   ` Will Deacon
2018-10-23  8:39   ` Ramana Radhakrishnan
2018-10-23  8:39     ` Ramana Radhakrishnan
2018-10-23  8:39     ` Ramana Radhakrishnan
2018-10-23  8:39     ` Ramana Radhakrishnan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20181019122204.GE14246@arm.com \
    --to=will.deacon@arm.com \
    --cc=Amit.Kachhap@arm.com \
    --cc=Christoffer.Dall@arm.com \
    --cc=Dave.Martin@arm.com \
    --cc=Jacob.Bramley@arm.com \
    --cc=Marc.Zyngier@arm.com \
    --cc=Mark.Rutland@arm.com \
    --cc=Suzuki.Poulose@arm.com \
    --cc=ard.biesheuvel@linaro.org \
    --cc=arnd@arndb.de \
    --cc=awallis@codeaurora.org \
    --cc=catalin.marinas@arm.com \
    --cc=drjones@redhat.com \
    --cc=kristina.martsenko@arm.com \
    --cc=kvmarm@lists.cs.columbia.edu \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ramana.radhakrishnan@foss.arm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.