netdev - Re: [net-next 1/1] tipc: avoid unnecessary copying of bundled messages

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: <20180214.152721.140431586969521284.davem@davemloft.net>
Date:   Wed, 14 Feb 2018 15:27:21 -0500 (EST)
From:   David Miller <davem@...emloft.net>
To:     jon.maloy@...csson.com
Cc:     netdev@...r.kernel.org,
        mohan.krishna.ghanta.krishnamurthy@...csson.com,
        tung.q.nguyen@...tech.com.au, hoang.h.le@...tech.com.au,
        canh.d.luu@...tech.com.au, ying.xue@...driver.com,
        tipc-discussion@...ts.sourceforge.net
Subject: Re: [net-next 1/1] tipc: avoid unnecessary copying of bundled
 messages

From: Jon Maloy <jon.maloy@...csson.com>
Date: Wed, 14 Feb 2018 13:50:31 +0100

> diff --git a/net/tipc/msg.c b/net/tipc/msg.c
> index 4e1c6f6..a368fa8 100644
> --- a/net/tipc/msg.c
> +++ b/net/tipc/msg.c
> @@ -434,6 +434,9 @@ bool tipc_msg_extract(struct sk_buff *skb, struct sk_buff **iskb, int *pos)
>  	skb_pull(*iskb, offset);
>  	imsz = msg_size(buf_msg(*iskb));
>  	skb_trim(*iskb, imsz);
> +
> +	/* Scale extracted buffer's truesize to avoid double accounting */
> +	(*iskb)->truesize = SKB_TRUESIZE(imsz);
>  	if (unlikely(!tipc_msg_validate(iskb)))
>  		goto none;
>  	*pos += align(imsz);

As Eric said, you have to be really careful here.

If you clone a 10K SKB 10 times, you really have to account for the full
truesize 10 times.

That is unless you explicitly trim off frags in the new clone, then adjust
the truesize by explicitly decreasing it by the amount of memory backing
the frags you trimmed off completely (not partially).

Finally, you can only do this on an SKB that has never entered a socket
SKB queue, otherwise you corrupt memory accounting.