linux-kernel - Re: [PATCH 2/3 v5] kernel/fork.c: avoid division by zero

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <alpine.DEB.2.10.1502241309570.3855@chino.kir.corp.google.com>
Date:	Tue, 24 Feb 2015 13:14:32 -0800 (PST)
From:	David Rientjes <rientjes@...gle.com>
To:	Heinrich Schuchardt <xypron.glpk@....de>
cc:	Andrew Morton <akpm@...ux-foundation.org>,
	Aaron Tomlin <atomlin@...hat.com>,
	Andy Lutomirski <luto@...capital.net>,
	Davidlohr Bueso <dave@...olabs.net>,
	"David S. Miller" <davem@...emloft.net>,
	Fabian Frederick <fabf@...net.be>,
	Guenter Roeck <linux@...ck-us.net>,
	"H. Peter Anvin" <hpa@...or.com>, Ingo Molnar <mingo@...nel.org>,
	Jens Axboe <axboe@...com>, Joe Perches <joe@...ches.com>,
	Johannes Weiner <hannes@...xchg.org>,
	Kees Cook <keescook@...omium.org>,
	Michael Marineau <mike@...ineau.org>,
	Oleg Nesterov <oleg@...hat.com>,
	"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
	Peter Zijlstra <peterz@...radead.org>,
	Prarit Bhargava <prarit@...hat.com>,
	Rik van Riel <riel@...hat.com>,
	Rusty Russell <rusty@...tcorp.com.au>,
	Steven Rostedt <rostedt@...dmis.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Vladimir Davydov <vdavydov@...allels.com>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/3 v5] kernel/fork.c: avoid division by zero

On Tue, 24 Feb 2015, Heinrich Schuchardt wrote:

> PAGE_SIZE is not guaranteed to be equal to or less than 8 times the
> THREAD_SIZE.
> 
> E.g. architecture hexagon may have page size 1M and thread size 4096.
> This would lead to a division by zero in the calculation of max_threads.
> 

This should only appear in one patch in the series, and I think this is 
the appropriate patch for that to happen.

> With 32-bit calculation there is no solution which delivers valid results
> for all possible combinations of the parameters.
> The code is only called once.
> Hence a 64-bit calculation can be used as solution.
> 
> Signed-off-by: Heinrich Schuchardt <xypron.glpk@....de>
> ---
>  kernel/fork.c | 38 +++++++++++++++++++++++++++-----------
>  1 file changed, 27 insertions(+), 11 deletions(-)
> 
> diff --git a/kernel/fork.c b/kernel/fork.c
> index 460b044..880c78d 100644
> --- a/kernel/fork.c
> +++ b/kernel/fork.c
> @@ -88,6 +88,16 @@
>  #include <trace/events/task.h>
>  
>  /*
> + * Minimum number of threads to boot the kernel
> + */
> +#define MIN_THREADS 20
> +
> +/*
> + * Maximum number of threads
> + */
> +#define MAX_THREADS FUTEX_TID_MASK
> +
> +/*
>   * Protected counters by write_lock_irq(&tasklist_lock)
>   */
>  unsigned long total_forks;	/* Handle normal Linux uptimes. */
> @@ -254,23 +264,29 @@ EXPORT_SYMBOL_GPL(__put_task_struct);
>  void __init __weak arch_task_cache_init(void) { }
>  
>  /*
> - * set_max_threads
> - * The argument is ignored.
> + * set_max_threads tries to set the default limit to the suggested value.

I'm not sure that is true.  At this point in the patch series, 
set_max_threads() is only called with UINT_MAX and that's not what the 
implementation tries to set max_threads to.

>   */
>  static void set_max_threads(unsigned int max_threads_suggested)
>  {
> -	/*
> -	 * The default maximum number of threads is set to a safe
> -	 * value: the thread structures can take up at most half
> -	 * of memory.
> -	 */
> -	max_threads = totalram_pages / (8 * THREAD_SIZE / PAGE_SIZE);
> +	u64 threads;
>  
>  	/*
> -	 * we need to allow at least 20 threads to boot a system
> +	 * The number of threads shall be limited such that the thread
> +	 * structures may only consume a small part of the available memory.
>  	 */
> -	if (max_threads < 20)
> -		max_threads = 20;
> +	threads = div64_u64((u64) totalram_pages * (u64) PAGE_SIZE,
> +			    (u64) THREAD_SIZE * 8UL);

The artithmetic works here, but I'm wondering what guarantee is in place 
to ensure that totalram_pages * PAGE_SIZE does not overflow on 64-bit 
arches?

> +
> +	if (threads > max_threads_suggested)
> +		threads = max_threads_suggested;

Ok, so now the function argument just shows an implementation issue.  
You're capping threads at UINT_MAX because you need max_threads to be an 
int and relying on the caller to enforce that.  set_max_threads() should 
be handling all of these details itself.

> +
> +	if (threads > MAX_THREADS)
> +		threads = MAX_THREADS;
> +
> +	if (threads < MIN_THREADS)
> +		threads = MIN_THREADS;
> +
> +	max_threads = (int) threads;
>  }
>  
>  void __init fork_init(void)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/