netdev - Re: [PATCH] bpf: Replace BUG_ON when fp

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: <20191216044948.gj7kwyfgr6laskip@ast-mbp.dhcp.thefacebook.com>
Date:   Sun, 15 Dec 2019 20:49:50 -0800
From:   Alexei Starovoitov <alexei.starovoitov@...il.com>
To:     Aditya Pakki <pakki001@....edu>
Cc:     kjlu@....edu, Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Martin KaFai Lau <kafai@...com>,
        Song Liu <songliubraving@...com>, Yonghong Song <yhs@...com>,
        Andrii Nakryiko <andriin@...com>, netdev@...r.kernel.org,
        bpf@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] bpf: Replace BUG_ON when fp_old is NULL

On Sun, Dec 15, 2019 at 09:44:32AM -0600, Aditya Pakki wrote:
> If fp_old is NULL in bpf_prog_realloc, the program does an assertion
> and crashes. However, we can continue execution by returning NULL to
> the upper callers. The patch fixes this issue.
> 
> Signed-off-by: Aditya Pakki <pakki001@....edu>
> ---
>  kernel/bpf/core.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
> index 49e32acad7d8..4b46654fb26b 100644
> --- a/kernel/bpf/core.c
> +++ b/kernel/bpf/core.c
> @@ -222,7 +222,8 @@ struct bpf_prog *bpf_prog_realloc(struct bpf_prog *fp_old, unsigned int size,
>  	u32 pages, delta;
>  	int ret;
>  
> -	BUG_ON(fp_old == NULL);
> +	if (!fp_old)
> +		return NULL;

This change makes no sense to me. fp_old should be valid. That's the point of
BUG_ON. It can happen only during development. Can remove BUG_ON just as well
and let kernel page fault.