All of lore.kernel.org
 help / color / mirror / Atom feed
From: Kees Cook <keescook@chromium.org>
To: kernel-hardening@lists.openwall.com
Cc: John Wood <john.wood@gmx.com>,
	Matthew Wilcox <willy@infradead.org>,
	Jonathan Corbet <corbet@lwn.net>,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	Ingo Molnar <mingo@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Juri Lelli <juri.lelli@redhat.com>,
	Vincent Guittot <vincent.guittot@linaro.org>,
	Dietmar Eggemann <dietmar.eggemann@arm.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
	Luis Chamberlain <mcgrof@kernel.org>,
	Iurii Zaikin <yzaikin@google.com>,
	James Morris <jmorris@namei.org>,
	"Serge E. Hallyn" <serge@hallyn.com>,
	linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-fsdevel@vger.kernel.org,
	linux-security-module@vger.kernel.org
Subject: Re: [RFC PATCH 4/6] security/fbfam: Add a new sysctl to control the crashing rate threshold
Date: Thu, 10 Sep 2020 16:14:38 -0700	[thread overview]
Message-ID: <202009101612.18BAD0241D@keescook> (raw)
In-Reply-To: <20200910202107.3799376-5-keescook@chromium.org>

On Thu, Sep 10, 2020 at 01:21:05PM -0700, Kees Cook wrote:
> From: John Wood <john.wood@gmx.com>
> 
> This is a previous step to add the detection feature.
> 
> A fork brute force attack will be detected when an application crashes
> quickly. Since, a rate can be defined as a time per fault, add a new
> sysctl to control the crashing rate threshold.
> 
> This way, each system can tune the detection's sensibility adjusting the
> milliseconds per fault. So, if the application's crashing rate falls
> under this threshold an attack will be detected. So, the higher this
> value, the faster an attack will be detected.
> 
> Signed-off-by: John Wood <john.wood@gmx.com>
> ---
>  include/fbfam/fbfam.h   |  4 ++++
>  kernel/sysctl.c         |  9 +++++++++
>  security/fbfam/Makefile |  1 +
>  security/fbfam/fbfam.c  | 11 +++++++++++
>  security/fbfam/sysctl.c | 20 ++++++++++++++++++++
>  5 files changed, 45 insertions(+)
>  create mode 100644 security/fbfam/sysctl.c
> 
> diff --git a/include/fbfam/fbfam.h b/include/fbfam/fbfam.h
> index b5b7d1127a52..2cfe51d2b0d5 100644
> --- a/include/fbfam/fbfam.h
> +++ b/include/fbfam/fbfam.h
> @@ -3,8 +3,12 @@
>  #define _FBFAM_H_
>  
>  #include <linux/sched.h>
> +#include <linux/sysctl.h>
>  
>  #ifdef CONFIG_FBFAM
> +#ifdef CONFIG_SYSCTL
> +extern struct ctl_table fbfam_sysctls[];
> +#endif

Instead of doing the extern and adding to sysctl.c, this can all be done
directly (dynamically) from the fbfam.c file instead.

>  int fbfam_fork(struct task_struct *child);
>  int fbfam_execve(void);
>  int fbfam_exit(void);
> diff --git a/kernel/sysctl.c b/kernel/sysctl.c
> index 09e70ee2332e..c3b4d737bef3 100644
> --- a/kernel/sysctl.c
> +++ b/kernel/sysctl.c
> @@ -77,6 +77,8 @@
>  #include <linux/uaccess.h>
>  #include <asm/processor.h>
>  
> +#include <fbfam/fbfam.h>
> +
>  #ifdef CONFIG_X86
>  #include <asm/nmi.h>
>  #include <asm/stacktrace.h>
> @@ -2660,6 +2662,13 @@ static struct ctl_table kern_table[] = {
>  		.extra1		= SYSCTL_ZERO,
>  		.extra2		= SYSCTL_ONE,
>  	},
> +#endif
> +#ifdef CONFIG_FBFAM
> +	{
> +		.procname	= "fbfam",
> +		.mode		= 0555,
> +		.child		= fbfam_sysctls,
> +	},
>  #endif
>  	{ }
>  };
> diff --git a/security/fbfam/Makefile b/security/fbfam/Makefile
> index f4b9f0b19c44..b8d5751ecea4 100644
> --- a/security/fbfam/Makefile
> +++ b/security/fbfam/Makefile
> @@ -1,2 +1,3 @@
>  # SPDX-License-Identifier: GPL-2.0
>  obj-$(CONFIG_FBFAM) += fbfam.o
> +obj-$(CONFIG_SYSCTL) += sysctl.o
> diff --git a/security/fbfam/fbfam.c b/security/fbfam/fbfam.c
> index 0387f95f6408..9be4639b72eb 100644
> --- a/security/fbfam/fbfam.c
> +++ b/security/fbfam/fbfam.c
> @@ -7,6 +7,17 @@
>  #include <linux/refcount.h>
>  #include <linux/slab.h>
>  
> +/**
> + * sysctl_crashing_rate_threshold - Crashing rate threshold.
> + *
> + * The rate's units are in milliseconds per fault.
> + *
> + * A fork brute force attack will be detected if the application's crashing rate
> + * falls under this threshold. So, the higher this value, the faster an attack
> + * will be detected.
> + */
> +unsigned long sysctl_crashing_rate_threshold = 30000;

I would move the sysctls here, instead. (Also, the above should be
const.)

> +
>  /**
>   * struct fbfam_stats - Fork brute force attack mitigation statistics.
>   * @refc: Reference counter.
> diff --git a/security/fbfam/sysctl.c b/security/fbfam/sysctl.c
> new file mode 100644
> index 000000000000..430323ad8e9f
> --- /dev/null
> +++ b/security/fbfam/sysctl.c
> @@ -0,0 +1,20 @@
> +// SPDX-License-Identifier: GPL-2.0
> +#include <linux/sysctl.h>
> +
> +extern unsigned long sysctl_crashing_rate_threshold;
> +static unsigned long ulong_one = 1;
> +static unsigned long ulong_max = ULONG_MAX;
> +
> +struct ctl_table fbfam_sysctls[] = {
> +	{
> +		.procname	= "crashing_rate_threshold",
> +		.data		= &sysctl_crashing_rate_threshold,
> +		.maxlen		= sizeof(sysctl_crashing_rate_threshold),
> +		.mode		= 0644,
> +		.proc_handler	= proc_doulongvec_minmax,
> +		.extra1		= &ulong_one,
> +		.extra2		= &ulong_max,
> +	},
> +	{ }
> +};

I wouldn't bother splitting this into a separate file. (Just leave it in
fbfam.c)

-- 
Kees Cook

  reply	other threads:[~2020-09-10 23:14 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-10 20:21 [RESEND][RFC PATCH 0/6] Fork brute force attack mitigation (fbfam) Kees Cook
2020-09-10 20:21 ` [RFC PATCH 1/6] security/fbfam: Add a Kconfig to enable the fbfam feature Kees Cook
2020-09-10 21:21   ` Jann Horn
2020-09-10 21:21     ` Jann Horn
2020-09-17 17:32     ` John Wood
2020-09-10 23:18   ` Kees Cook
2020-09-17 18:40     ` John Wood
2020-09-17 22:05       ` Kees Cook
2020-09-18 14:50         ` John Wood
2020-09-10 20:21 ` [RFC PATCH 2/6] security/fbfam: Add the api to manage statistics Kees Cook
2020-09-10 23:23   ` Kees Cook
2020-09-10 20:21 ` [RFC PATCH 3/6] security/fbfam: Use " Kees Cook
2020-09-10 20:27   ` Jann Horn
2020-09-10 20:27     ` Jann Horn
2020-09-10 23:33   ` Kees Cook
2020-09-29 23:47     ` Steven Rostedt
2020-09-29 23:49       ` Steven Rostedt
2020-10-03  9:52         ` John Wood
2020-09-10 20:21 ` [RFC PATCH 4/6] security/fbfam: Add a new sysctl to control the crashing rate threshold Kees Cook
2020-09-10 23:14   ` Kees Cook [this message]
2020-09-13 14:33     ` John Wood
2020-09-10 20:21 ` [RFC PATCH 5/6] security/fbfam: Detect a fork brute force attack Kees Cook
2020-09-10 21:10   ` Jann Horn
2020-09-10 21:10     ` Jann Horn
2020-09-13 17:54     ` John Wood
2020-09-14 19:42       ` Jann Horn
2020-09-14 19:42         ` Jann Horn
2020-09-15 18:44         ` John Wood
2020-09-10 23:49   ` Kees Cook
2020-09-11  0:01     ` Jann Horn
2020-09-11  0:01       ` Jann Horn
2020-09-13 16:56       ` John Wood
2020-09-14 19:39         ` Jann Horn
2020-09-14 19:39           ` Jann Horn
2020-09-15 17:36           ` John Wood
2020-09-10 20:21 ` [RFC PATCH 6/6] security/fbfam: Mitigate " Kees Cook
2020-09-10 20:55   ` Jann Horn
2020-09-10 20:55     ` Jann Horn
2020-09-10 23:56   ` Kees Cook
2020-09-11  0:20     ` Jann Horn
2020-09-11  0:20       ` Jann Horn
2020-09-18 16:02     ` John Wood
2020-09-18 21:35       ` Kees Cook
2020-09-19  8:01         ` John Wood
2020-09-10 20:39 ` [RESEND][RFC PATCH 0/6] Fork brute force attack mitigation (fbfam) Jann Horn
2020-09-10 20:39   ` Jann Horn
2020-09-10 23:58 ` Kees Cook
2020-09-11 14:48   ` John Wood
2020-09-12  7:55     ` Kees Cook
2020-09-12 12:24       ` John Wood
2020-09-12  0:03 ` James Morris
2020-09-12  7:56   ` Kees Cook
2020-09-12  9:36     ` John Wood
2020-09-12 14:47       ` Mel Gorman
2020-09-12 20:48         ` Ondrej Mosnacek
2020-09-12 20:48           ` Ondrej Mosnacek
2020-09-13  7:24           ` John Wood
2020-09-13  7:24             ` John Wood

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=202009101612.18BAD0241D@keescook \
    --to=keescook@chromium.org \
    --cc=bsegall@google.com \
    --cc=corbet@lwn.net \
    --cc=dietmar.eggemann@arm.com \
    --cc=jmorris@namei.org \
    --cc=john.wood@gmx.com \
    --cc=juri.lelli@redhat.com \
    --cc=kernel-hardening@lists.openwall.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-security-module@vger.kernel.org \
    --cc=mcgrof@kernel.org \
    --cc=mgorman@suse.de \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=serge@hallyn.com \
    --cc=vincent.guittot@linaro.org \
    --cc=viro@zeniv.linux.org.uk \
    --cc=willy@infradead.org \
    --cc=yzaikin@google.com \
    --subject='Re: [RFC PATCH 4/6] security/fbfam: Add a new sysctl to control the crashing rate threshold' \
    /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

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.