Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Fri, 26 Jan 2018 23:01:21 +0100 (CET)
From:   Jiri Kosina <jikos@kernel.org>
To:     Kees Cook <keescook@chromium.org>
cc:     Andrew Morton <akpm@linux-foundation.org>,
        Andy Lutomirski <luto@kernel.org>,
        Jann Horn <jannh@google.com>, Ingo Molnar <mingo@kernel.org>,
        Laura Abbott <labbott@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Al Viro <viro@zeniv.linux.org.uk>,
        Sahara <keun-o.park@darkmatter.ae>,
        "Levin, Alexander (Sasha Levin)" <alexander.levin@verizon.com>,
        Michal Hocko <mhocko@suse.com>,
        Andrea Arcangeli <aarcange@redhat.com>,
        "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
        linux-kernel@vger.kernel.org, kernel-hardening@lists.openwall.com
Subject: Re: [PATCH] fork: Allow stack to be wiped on fork
In-Reply-To: <20180117055015.GA15256@beast>
Message-ID: <alpine.LRH.2.00.1801262259300.27010@gjva.wvxbf.pm>
References: <20180117055015.GA15256@beast>
User-Agent: Alpine 2.00 (LRH 1167 2008-08-23)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, 16 Jan 2018, Kees Cook wrote:

> diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h
> index 34f053a150a9..091f53fe31cc 100644
> --- a/include/linux/thread_info.h
> +++ b/include/linux/thread_info.h
> @@ -43,7 +43,9 @@ enum {
>  #define THREAD_ALIGN	THREAD_SIZE
>  #endif
>  
> -#if IS_ENABLED(CONFIG_DEBUG_STACK_USAGE) || IS_ENABLED(CONFIG_DEBUG_KMEMLEAK)
> +#if IS_ENABLED(CONFIG_DEBUG_STACK_USAGE) || \
> +    IS_ENABLED(CONFIG_DEBUG_KMEMLEAK) || \
> +    IS_ENABLED(CONFIG_CLEAR_STACK_FORK)
>  # define THREADINFO_GFP		(GFP_KERNEL_ACCOUNT | __GFP_ZERO)
>  #else
>  # define THREADINFO_GFP		(GFP_KERNEL_ACCOUNT)
> diff --git a/kernel/fork.c b/kernel/fork.c
> index 2295fc69717f..215b1ce2b2cd 100644
> --- a/kernel/fork.c
> +++ b/kernel/fork.c
> @@ -215,7 +215,7 @@ static unsigned long *alloc_thread_stack_node(struct task_struct *tsk, int node)
>  		if (!s)
>  			continue;
>  
> -#ifdef CONFIG_DEBUG_KMEMLEAK
> +#if IS_ENABLED(CONFIG_DEBUG_KMEMLEAK) || IS_ENABLED(CONFIG_CLEAR_STACK_FORK)
>  		/* Clear stale pointers from reused stack. */
>  		memset(s->addr, 0, THREAD_SIZE);
>  #endif

Is there any good reason not to do it symmetricaly also for non-vmapped 
stacks? (by passing __GFP_ZERO to alloc_pages_node())?

Thanks,

-- 
Jiri Kosina
SUSE Labs