From: Will Deacon <will@kernel.org>
To: Marco Elver <elver@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>,
Kees Cook <keescook@chromium.org>,
"Paul E. McKenney" <paulmck@kernel.org>,
"Michael S. Tsirkin" <mst@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Jason Wang <jasowang@redhat.com>,
Nick Desaulniers <ndesaulniers@google.com>,
Josh Triplett <josh@joshtriplett.org>,
Arnd Bergmann <arnd@arndb.de>,
LKML <linux-kernel@vger.kernel.org>,
Ivan Kokshaysky <ink@jurassic.park.msu.ru>,
Sami Tolvanen <samitolvanen@google.com>,
linux-alpha@vger.kernel.org,
Alan Stern <stern@rowland.harvard.edu>,
Matt Turner <mattst88@gmail.com>,
virtualization@lists.linux-foundation.org,
Android Kernel Team <kernel-team@android.com>,
Boqun Feng <boqun.feng@gmail.com>,
linux-arm-kernel@lists.infradead.org,
Richard Henderson <rth@twiddle.net>
Subject: Re: [PATCH 18/18] arm64: lto: Strengthen READ_ONCE() to acquire when CLANG_LTO=y
Date: Wed, 1 Jul 2020 11:24:28 +0100 [thread overview]
Message-ID: <20200701102427.GD14959@willie-the-truck> (raw)
In-Reply-To: <CANpmjNPOO=AVsVJMdL8sq03jwHsDR_1_FfWccwaLKEBRn1RFtA@mail.gmail.com>
On Tue, Jun 30, 2020 at 09:47:30PM +0200, Marco Elver wrote:
> On Tue, 30 Jun 2020 at 19:39, Will Deacon <will@kernel.org> wrote:
> >
> > When building with LTO, there is an increased risk of the compiler
> > converting an address dependency headed by a READ_ONCE() invocation
> > into a control dependency and consequently allowing for harmful
> > reordering by the CPU.
> >
> > Ensure that such transformations are harmless by overriding the generic
> > READ_ONCE() definition with one that provides acquire semantics when
> > building with LTO.
> >
> > Signed-off-by: Will Deacon <will@kernel.org>
> > ---
> > arch/arm64/include/asm/rwonce.h | 63 +++++++++++++++++++++++++++++++
> > arch/arm64/kernel/vdso/Makefile | 2 +-
> > arch/arm64/kernel/vdso32/Makefile | 2 +-
> > 3 files changed, 65 insertions(+), 2 deletions(-)
> > create mode 100644 arch/arm64/include/asm/rwonce.h
>
> This seems reasonable, given we can't realistically tell the compiler
> about dependent loads. What (if any), is the performance impact? I
> guess this also heavily depends on the actual silicon.
Right, it depends both on the CPU micro-architecture and also the workload.
When we ran some basic tests, the overhead wasn't greater than the benefit
seen by enabling LTO, so it seems like a reasonable trade-off (given that
LTO is a dependency for CFI, so it's not just about performance).
Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2020-07-01 10:25 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-30 17:37 [PATCH 00/18] Allow architectures to override __READ_ONCE() Will Deacon
2020-06-30 17:37 ` [PATCH 01/18] tools: bpf: Use local copy of headers including uapi/linux/filter.h Will Deacon
2020-07-01 16:38 ` Alexei Starovoitov
2020-06-30 17:37 ` [PATCH 02/18] compiler.h: Split {READ, WRITE}_ONCE definitions out into rwonce.h Will Deacon
2020-06-30 19:11 ` Arnd Bergmann
2020-07-01 10:16 ` [PATCH 02/18] compiler.h: Split {READ,WRITE}_ONCE " Will Deacon
2020-07-01 11:33 ` [PATCH 02/18] compiler.h: Split {READ, WRITE}_ONCE " Arnd Bergmann
2020-06-30 17:37 ` [PATCH 03/18] asm/rwonce: Allow __READ_ONCE to be overridden by the architecture Will Deacon
2020-06-30 17:37 ` [PATCH 04/18] alpha: Override READ_ONCE() with barriered implementation Will Deacon
2020-07-02 9:32 ` Mark Rutland
2020-07-02 9:48 ` Will Deacon
2020-07-02 10:08 ` Arnd Bergmann
2020-07-02 11:18 ` Will Deacon
2020-07-02 11:39 ` Arnd Bergmann
2020-07-02 14:43 ` Joel Fernandes
2020-07-02 14:55 ` Will Deacon
2020-07-02 15:07 ` Joel Fernandes
2020-06-30 17:37 ` [PATCH 05/18] asm/rwonce: Remove smp_read_barrier_depends() invocation Will Deacon
2020-06-30 17:37 ` [PATCH 06/18] vhost: Remove redundant use of read_barrier_depends() barrier Will Deacon
2020-06-30 17:37 ` [PATCH 07/18] alpha: Replace smp_read_barrier_depends() usage with smp_[r]mb() Will Deacon
2020-06-30 17:37 ` [PATCH 08/18] locking/barriers: Remove definitions for [smp_]read_barrier_depends() Will Deacon
2020-06-30 17:37 ` [PATCH 09/18] Documentation/barriers: Remove references to [smp_]read_barrier_depends() Will Deacon
2020-06-30 17:37 ` [PATCH 10/18] Documentation/barriers/kokr: " Will Deacon
2020-06-30 17:37 ` [PATCH 11/18] tools/memory-model: Remove smp_read_barrier_depends() from informal doc Will Deacon
2020-06-30 17:37 ` [PATCH 12/18] include/linux: Remove smp_read_barrier_depends() from comments Will Deacon
2020-06-30 17:37 ` [PATCH 13/18] checkpatch: Remove checks relating to [smp_]read_barrier_depends() Will Deacon
2020-06-30 17:37 ` [PATCH 14/18] arm64: Reduce the number of header files pulled into vmlinux.lds.S Will Deacon
2020-06-30 17:37 ` [PATCH 15/18] arm64: alternatives: Split up alternative.h Will Deacon
2020-06-30 17:37 ` [PATCH 16/18] arm64: cpufeatures: Add capability for LDAPR instruction Will Deacon
2020-06-30 17:37 ` [PATCH 17/18] arm64: alternatives: Remove READ_ONCE() usage during patch operation Will Deacon
2020-06-30 17:37 ` [PATCH 18/18] arm64: lto: Strengthen READ_ONCE() to acquire when CLANG_LTO=y Will Deacon
2020-06-30 19:25 ` Arnd Bergmann
2020-07-01 10:19 ` Will Deacon
2020-07-01 10:59 ` Arnd Bergmann
2020-06-30 19:47 ` Marco Elver
2020-06-30 20:20 ` Peter Zijlstra
2020-06-30 22:57 ` Sami Tolvanen
2020-07-01 10:25 ` Will Deacon
2020-07-01 10:24 ` Will Deacon [this message]
2020-07-01 17:07 ` Dave P Martin
2020-07-02 7:23 ` Will Deacon
2020-07-06 16:00 ` Dave Martin
2020-07-06 16:34 ` Paul E. McKenney
2020-07-06 17:05 ` Dave Martin
2020-07-06 17:36 ` Paul E. McKenney
2020-07-07 10:29 ` Dave Martin
2020-07-07 22:51 ` Paul E. McKenney
2020-07-07 23:01 ` Nick Desaulniers
2020-07-08 7:15 ` Marco Elver
2020-07-08 9:16 ` Peter Zijlstra
2020-07-08 18:20 ` Paul E. McKenney
2020-07-06 18:35 ` Will Deacon
2020-07-06 19:23 ` Marco Elver
2020-07-06 19:42 ` Paul E. McKenney
2020-07-06 16:08 ` Dave Martin
2020-07-06 18:35 ` Will Deacon
2020-07-07 10:10 ` Dave Martin
2020-07-01 7:38 ` [PATCH 00/18] Allow architectures to override __READ_ONCE() Josh Triplett
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=20200701102427.GD14959@willie-the-truck \
--to=will@kernel.org \
--cc=arnd@arndb.de \
--cc=boqun.feng@gmail.com \
--cc=catalin.marinas@arm.com \
--cc=elver@google.com \
--cc=ink@jurassic.park.msu.ru \
--cc=jasowang@redhat.com \
--cc=josh@joshtriplett.org \
--cc=keescook@chromium.org \
--cc=kernel-team@android.com \
--cc=linux-alpha@vger.kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=mattst88@gmail.com \
--cc=mst@redhat.com \
--cc=ndesaulniers@google.com \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=rth@twiddle.net \
--cc=samitolvanen@google.com \
--cc=stern@rowland.harvard.edu \
--cc=virtualization@lists.linux-foundation.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).