linux-kernel - Re: [PATCH 1/1] seccomp: release seccomp after killing terminate clone/fork

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 for Android: free password hash cracker in your pocket

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

Date:   Wed, 29 Dec 2021 15:53:05 -0800
From:   Andrew Morton <akpm@...ux-foundation.org>
To:     Calvin Liao <calvin.liao@...il.com>
Cc:     Matthias Brugger <matthias.bgg@...il.com>,
        "Eric W . Biederman" <ebiederm@...ssion.com>,
        "Peter Zijlstra" <peterz@...radead.org>,
        Christian Brauner <christian.brauner@...ntu.com>,
        Jens Axboe <axboe@...nel.dk>,
        Alexey Gladkov <legion@...nel.org>,
        Mike Christie <michael.christie@...cle.com>,
        "David Hildenbrand" <david@...hat.com>,
        Rolf Eike Beer <eb@...ix.com>,
        <linux-kernel@...r.kernel.org>,
        <linux-arm-kernel@...ts.infradead.org>,
        <linux-mediatek@...ts.infradead.org>,
        wsd_upstream <wsd_upstream@...iatek.com>,
        "calvin.liao" <calvin.liao@...iatek.com>,
        Andrea Arcangeli <aarcange@...hat.com>
Subject: Re: [PATCH 1/1] seccomp: release seccomp after killing terminate
 clone/fork

On Tue, 28 Dec 2021 13:34:21 +0800 Calvin Liao <calvin.liao@...il.com> wrote:

> From: "calvin.liao" <calvin.liao@...iatek.com>
> 
> Add to release seccomp after killing terminate clone/fork to avoid
> memory leak when enabling CONFIG_SECCOMP.
> 

Help us out here - where was this filter allocated?  Please describe
the code flow which led to this leak.  Rather than simply saying "there
is a leak".

> --- a/kernel/fork.c
> +++ b/kernel/fork.c
> @@ -1698,6 +1698,13 @@ static void copy_seccomp(struct task_struct *p)
>  #endif
>  }
>  
> +static void release_seccomp(struct task_struct *p)
> +{
> +#ifdef CONFIG_SECCOMP
> +	seccomp_filter_release(p);
> +#endif

ifdefs aren't needed?

> +}
> +
>  SYSCALL_DEFINE1(set_tid_address, int __user *, tidptr)
>  {
>  	current->clear_child_tid = tidptr;
> @@ -2405,6 +2412,7 @@ static __latent_entropy struct task_struct *copy_process(
>  	return p;
>  
>  bad_fork_cancel_cgroup:
> +	release_seccomp(p);

Why not simply call seccomp_filter_release() here, without ifdefs?

>  	sched_core_free(p);
>  	spin_unlock(&current->sighand->siglock);
>  	write_unlock_irq(&tasklist_lock);