From: Andrey Konovalov <andreyknvl@google.com>
To: Andrew Morton <akpm@linux-foundation.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Vincenzo Frascino <vincenzo.frascino@arm.com>,
Dmitry Vyukov <dvyukov@google.com>,
Alexander Potapenko <glider@google.com>,
Marco Elver <elver@google.com>
Cc: Will Deacon <will.deacon@arm.com>,
Andrey Ryabinin <aryabinin@virtuozzo.com>,
Peter Collingbourne <pcc@google.com>,
Evgenii Stepanov <eugenis@google.com>,
Branislav Rankov <Branislav.Rankov@arm.com>,
Kevin Brodsky <kevin.brodsky@arm.com>,
kasan-dev <kasan-dev@googlegroups.com>,
Linux ARM <linux-arm-kernel@lists.infradead.org>,
Linux Memory Management List <linux-mm@kvack.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 00/12] kasan: optimizations and fixes for HW_TAGS
Date: Fri, 5 Feb 2021 17:25:44 +0100 [thread overview]
Message-ID: <CAAeHK+zv7U_oN1WVqQNhorL4Gf9G-hFb120o3XFO9RDtY7TEpQ@mail.gmail.com> (raw)
In-Reply-To: <cover.1612538932.git.andreyknvl@google.com>
On Fri, Feb 5, 2021 at 4:39 PM Andrey Konovalov <andreyknvl@google.com> wrote:
>
> This patchset goes on top of:
>
> 1. Vincenzo's async support patches [1], and
> 2. "kasan: untag addresses for KFENCE" fix [2] (already in mm).
>
> [1] https://lore.kernel.org/linux-arm-kernel/20210130165225.54047-1-vincenzo.frascino@arm.com/
> [2] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=akpm&id=dec4728fab910da0c86cf9a97e980f4244ebae9f
>
> This patchset makes the HW_TAGS mode more efficient, mostly by reworking
> poisoning approaches and simplifying/inlining some internal helpers.
>
> With this change, the overhead of HW_TAGS annotations excluding setting
> and checking memory tags is ~3%. The performance impact caused by tags
> will be unknown until we have hardware that supports MTE.
>
> As a side-effect, this patchset speeds up generic KASAN by ~15%.
Forgot to include changes v1->v2:
- Use EXPORT_SYMBOL_GPL() for arm64 symbols.
- Rename kmalloc bool flag argument to is_kmalloc.
- Make empty mte_set_mem_tag_range() return void.
- Fix build warning in 32-bit systems by using unsigned long instead
of u64 in WARN_ON() checks.
- Minor changes in comments and commit descriptions.
- Use kfence_ksize() before __ksize() to avoid crashes with KFENCE.
- Use inline instead of __always_inline.
>
> Andrey Konovalov (12):
> kasan, mm: don't save alloc stacks twice
> kasan, mm: optimize kmalloc poisoning
> kasan: optimize large kmalloc poisoning
> kasan: clean up setting free info in kasan_slab_free
> kasan: unify large kfree checks
> kasan: rework krealloc tests
> kasan, mm: fail krealloc on freed objects
> kasan, mm: optimize krealloc poisoning
> kasan: ensure poisoning size alignment
> arm64: kasan: simplify and inline MTE functions
> kasan: inline HW_TAGS helper functions
> arm64: kasan: export MTE symbols for KASAN tests
>
> arch/arm64/include/asm/cache.h | 1 -
> arch/arm64/include/asm/kasan.h | 1 +
> arch/arm64/include/asm/mte-def.h | 2 +
> arch/arm64/include/asm/mte-kasan.h | 65 ++++++++--
> arch/arm64/include/asm/mte.h | 2 -
> arch/arm64/kernel/mte.c | 48 +-------
> arch/arm64/lib/mte.S | 16 ---
> include/linux/kasan.h | 25 ++--
> lib/test_kasan.c | 111 +++++++++++++++--
> mm/kasan/common.c | 187 ++++++++++++++++++++---------
> mm/kasan/kasan.h | 72 +++++++++--
> mm/kasan/shadow.c | 53 ++++----
> mm/slab_common.c | 18 ++-
> mm/slub.c | 3 +-
> 14 files changed, 418 insertions(+), 186 deletions(-)
>
> --
> 2.30.0.365.g02bc693789-goog
>
next prev parent reply other threads:[~2021-02-05 22:06 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-05 15:39 [PATCH v2 00/12] kasan: optimizations and fixes for HW_TAGS Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 01/12] kasan, mm: don't save alloc stacks twice Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 02/12] kasan, mm: optimize kmalloc poisoning Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 03/12] kasan: optimize large " Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 04/12] kasan: clean up setting free info in kasan_slab_free Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 05/12] kasan: unify large kfree checks Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 06/12] kasan: rework krealloc tests Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 07/12] kasan, mm: fail krealloc on freed objects Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 08/12] kasan, mm: optimize krealloc poisoning Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 09/12] kasan: ensure poisoning size alignment Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 10/12] arm64: kasan: simplify and inline MTE functions Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 11/12] kasan: inline HW_TAGS helper functions Andrey Konovalov
2021-02-05 15:39 ` [PATCH v2 12/12] arm64: kasan: export MTE symbols for KASAN tests Andrey Konovalov
2021-02-08 17:42 ` Christoph Hellwig
2021-02-09 15:18 ` Andrey Konovalov
2021-02-05 16:25 ` Andrey Konovalov [this message]
2021-02-05 16:53 ` [PATCH v2 00/12] kasan: optimizations and fixes for HW_TAGS Andrey Konovalov
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=CAAeHK+zv7U_oN1WVqQNhorL4Gf9G-hFb120o3XFO9RDtY7TEpQ@mail.gmail.com \
--to=andreyknvl@google.com \
--cc=Branislav.Rankov@arm.com \
--cc=akpm@linux-foundation.org \
--cc=aryabinin@virtuozzo.com \
--cc=catalin.marinas@arm.com \
--cc=dvyukov@google.com \
--cc=elver@google.com \
--cc=eugenis@google.com \
--cc=glider@google.com \
--cc=kasan-dev@googlegroups.com \
--cc=kevin.brodsky@arm.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=pcc@google.com \
--cc=vincenzo.frascino@arm.com \
--cc=will.deacon@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 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).