linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Brendan Higgins <brendanhiggins@google.com>
To: Patricia Alfonso <trishalfonso@google.com>
Cc: Jeff Dike <jdike@addtoit.com>,
	Richard Weinberger <richard@nod.at>,
	Anton Ivanov <anton.ivanov@cambridgegreys.com>,
	aryabinin@virtuozzo.com, Dmitry Vyukov <dvyukov@google.com>,
	David Gow <davidgow@google.com>,
	Johannes Berg <johannes@sipsolutions.net>,
	kasan-dev@googlegroups.com,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	linux-um <linux-um@lists.infradead.org>
Subject: Re: [PATCH] UML: add support for KASAN under x86_64
Date: Tue, 25 Feb 2020 17:19:21 -0800	[thread overview]
Message-ID: <CAFd5g45gqZcJ6v3KSDuBffgBzfZ+=GJ2oCuSurYehoMHBK0Grg@mail.gmail.com> (raw)
In-Reply-To: <20200226004608.8128-1-trishalfonso@google.com>

On Tue, Feb 25, 2020 at 4:46 PM Patricia Alfonso
<trishalfonso@google.com> wrote:
>
> Make KASAN run on User Mode Linux on x86_64.
>
> Depends on Constructor support in UML - "[RFC PATCH] um:
> implement CONFIG_CONSTRUCTORS for modules"
> (https://patchwork.ozlabs.org/patch/1234551/) by Johannes Berg.
>
> The location of the KASAN shadow memory, starting at
> KASAN_SHADOW_OFFSET, can be configured using the
> KASAN_SHADOW_OFFSET option. UML uses roughly 18TB of address
> space, and KASAN requires 1/8th of this. The default location of
> this offset is 0x7fff8000 as suggested by Dmitry Vyukov. There is
> usually enough free space at this location; however, it is a config
> option so that it can be easily changed if needed.
>
> The UML-specific KASAN initializer uses mmap to map
> the roughly 2.25TB of shadow memory to the location defined by
> KASAN_SHADOW_OFFSET. kasan_init() utilizes constructors to initialize
> KASAN before main().
>
> Disable stack instrumentation on UML via KASAN_STACK config option to
> avoid false positive KASAN reports.
>
> Signed-off-by: Patricia Alfonso <trishalfonso@google.com>

A couple of minor nits (well one nit and one question), but overall
this looks good to me.

Reviewed-by: Brendan Higgins <brendanhiggins@google.com>

> ---
>  arch/um/Kconfig                  | 13 +++++++++++++
>  arch/um/Makefile                 |  6 ++++++
>  arch/um/include/asm/common.lds.S |  1 +
>  arch/um/include/asm/kasan.h      | 32 ++++++++++++++++++++++++++++++++
>  arch/um/kernel/dyn.lds.S         |  5 ++++-
>  arch/um/kernel/mem.c             | 18 ++++++++++++++++++
>  arch/um/os-Linux/mem.c           | 22 ++++++++++++++++++++++
>  arch/um/os-Linux/user_syms.c     |  4 ++--
>  arch/x86/um/Makefile             |  3 ++-
>  arch/x86/um/vdso/Makefile        |  3 +++
>  lib/Kconfig.kasan                |  2 +-
>  11 files changed, 104 insertions(+), 5 deletions(-)
>  create mode 100644 arch/um/include/asm/kasan.h
>
> diff --git a/arch/um/Kconfig b/arch/um/Kconfig
> index 0917f8443c28..fb2ad1fb05fd 100644
> --- a/arch/um/Kconfig
> +++ b/arch/um/Kconfig
> @@ -8,6 +8,7 @@ config UML
>         select ARCH_HAS_KCOV
>         select ARCH_NO_PREEMPT
>         select HAVE_ARCH_AUDITSYSCALL
> +       select HAVE_ARCH_KASAN if X86_64
>         select HAVE_ARCH_SECCOMP_FILTER
>         select HAVE_ASM_MODVERSIONS
>         select HAVE_UID16
> @@ -200,6 +201,18 @@ config UML_TIME_TRAVEL_SUPPORT
>
>           It is safe to say Y, but you probably don't need this.
>
> +config KASAN_SHADOW_OFFSET
> +       hex
> +       depends on KASAN
> +       default 0x7fff8000

nit: It looks like you chose the default that Dmitry suggested. Some
explanation of this in the help would probably be good.

> +       help
> +         This is the offset at which the ~2.25TB of shadow memory is
> +         mapped and used by KASAN for memory debugging. This can be any
> +         address that has at least KASAN_SHADOW_SIZE(total address space divided
> +         by 8) amount of space so that the KASAN shadow memory does not conflict
> +         with anything. The default is 0x7fff8000, as it fits into immediate of
> +         most instructions.
> +
>  endmenu

[...]

> diff --git a/lib/Kconfig.kasan b/lib/Kconfig.kasan
> index 81f5464ea9e1..5b54f3c9a741 100644
> --- a/lib/Kconfig.kasan
> +++ b/lib/Kconfig.kasan
> @@ -125,7 +125,7 @@ config KASAN_STACK_ENABLE
>
>  config KASAN_STACK
>         int
> -       default 1 if KASAN_STACK_ENABLE || CC_IS_GCC
> +       default 1 if (KASAN_STACK_ENABLE || CC_IS_GCC) && !UML

Up to the KASAN people, but I think you can probably move this to
arch/um/Kconfig. There is some advantage to having all the UML
specific Kconfigery in arch/um/Kconfig, but there are also already a
lot of things that specify !UML outside of arch/um/.

>         default 0
>
>  config KASAN_S390_4_LEVEL_PAGING
> --
> 2.25.0.265.gbab2e86ba0-goog
>

  reply	other threads:[~2020-02-26  1:19 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-26  0:46 [PATCH] UML: add support for KASAN under x86_64 Patricia Alfonso
2020-02-26  1:19 ` Brendan Higgins [this message]
2020-02-26 15:24 ` Dmitry Vyukov
2020-03-06  0:03 ` Patricia Alfonso
2020-03-11 10:32   ` Johannes Berg
2020-03-11 10:46     ` Dmitry Vyukov
2020-03-11 11:18     ` Johannes Berg
2020-03-11 11:40       ` Johannes Berg
2020-03-11 17:34       ` Dmitry Vyukov
2020-03-20 13:39         ` Johannes Berg
2020-03-20 15:18           ` Dmitry Vyukov
2020-03-30  7:43             ` Johannes Berg
2020-03-30  8:38               ` Dmitry Vyukov
2020-03-30  8:41                 ` Johannes Berg
2020-03-31  6:14                   ` David Gow
2020-03-31  7:43                     ` Johannes Berg
2020-03-31 16:39                   ` Patricia Alfonso
2020-03-31 16:54                     ` Richard Weinberger
2020-03-11 22:32     ` Patricia Alfonso
2020-03-11 22:44       ` Johannes Berg
2022-05-24 10:34         ` Vincent Whitchurch
2022-05-24 10:45           ` Johannes Berg
2022-05-24 19:35           ` David Gow
2022-05-25 11:17             ` Vincent Whitchurch
2022-05-26  1:01               ` [RFC PATCH v3] " David Gow
2022-05-26  9:29                 ` Johannes Berg
2022-05-27  5:31                 ` Dmitry Vyukov
2022-05-27  7:32                   ` Johannes Berg
2022-05-27 10:36                 ` Johannes Berg
2022-05-27 13:05                 ` Johannes Berg
2022-05-27 13:09                   ` Dmitry Vyukov
2022-05-27 13:15                     ` Johannes Berg
2022-05-27 13:18                       ` Dmitry Vyukov
2022-05-27 13:27                         ` Johannes Berg
2022-05-27 13:52                           ` Dmitry Vyukov
2022-05-27 14:27                             ` Johannes Berg
2022-05-27 15:46                               ` Dmitry Vyukov
2020-03-29 19:06     ` [PATCH] " Richard Weinberger

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='CAFd5g45gqZcJ6v3KSDuBffgBzfZ+=GJ2oCuSurYehoMHBK0Grg@mail.gmail.com' \
    --to=brendanhiggins@google.com \
    --cc=anton.ivanov@cambridgegreys.com \
    --cc=aryabinin@virtuozzo.com \
    --cc=davidgow@google.com \
    --cc=dvyukov@google.com \
    --cc=jdike@addtoit.com \
    --cc=johannes@sipsolutions.net \
    --cc=kasan-dev@googlegroups.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-um@lists.infradead.org \
    --cc=richard@nod.at \
    --cc=trishalfonso@google.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).