netdev - Re: [PATCH v2 net-next] tcp: md5: use kmalloc() backed scratch areas

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]

Date:	Mon, 27 Jun 2016 11:31:01 -0700
From:	Cong Wang <xiyou.wangcong@...il.com>
To:	Eric Dumazet <eric.dumazet@...il.com>
Cc:	Andy Lutomirski <luto@...capital.net>,
	David Miller <davem@...emloft.net>,
	Herbert Xu <herbert@...dor.apana.org.au>,
	Network Development <netdev@...r.kernel.org>
Subject: Re: [PATCH v2 net-next] tcp: md5: use kmalloc() backed scratch areas

On Mon, Jun 27, 2016 at 9:51 AM, Eric Dumazet <eric.dumazet@...il.com> wrote:
> diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
> index 5c7ed147449c1b7ba029b12e033ad779a631460a..fddc0ab799996c1df82cb05dba03271b773e3b2d 100644
> --- a/net/ipv4/tcp.c
> +++ b/net/ipv4/tcp.c
> @@ -2969,8 +2969,18 @@ static void __tcp_alloc_md5sig_pool(void)
>                 return;
>
>         for_each_possible_cpu(cpu) {
> +               void *scratch = per_cpu(tcp_md5sig_pool, cpu).scratch;
>                 struct ahash_request *req;
>
> +               if (!scratch) {
> +                       scratch = kmalloc_node(sizeof(union tcp_md5sum_block) +
> +                                              sizeof(struct tcphdr),
> +                                              GFP_KERNEL,
> +                                              cpu_to_node(cpu));
> +                       if (!scratch)
> +                               return;
> +                       per_cpu(tcp_md5sig_pool, cpu).scratch = scratch;
> +               }
>                 if (per_cpu(tcp_md5sig_pool, cpu).md5_req)
>                         continue;

Not a problem of your patch, but it seems these allocations never
get freed once we start using tcp md5. Maybe we should free them
when the last socket using tcp md5 is gone?