netdev - Re: [PATCH] net: hdlc: Increase maximum HDLC MTU

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: <Y7W84D+J4iNx30zx@lunn.ch>
Date:   Wed, 4 Jan 2023 18:52:32 +0100
From:   Andrew Lunn <andrew@...n.ch>
To:     Harald Welte <laforge@...ocom.org>
Cc:     netdev@...r.kernel.org, khc@...waw.pl
Subject: Re: [PATCH] net: hdlc: Increase maximum HDLC MTU

On Wed, Jan 04, 2023 at 01:57:24PM +0100, Harald Welte wrote:
> The FRF 1.2 specification clearly states:
> 
> > A maximum frame relay information field size of 1600 octets shall be
> > supported by the network and the user.
> 
> The linux kernel hdlc/fr code has so far had a maximum MTU of 1500
> octets.  This may have been sufficient to transport "regular" Ethernet
> frames of MTU 1500 via frame relay net-devices, but there are other
> use cases than Ethernet.
> 
> One such use case is the 3GPP Gb interface (TS 48.014, 48.016, 48.018)
> operated over Frame Relay.  There is open source software [2]
> implementing those interfaces by means of AF_PACKET sockets over
> Linux kernel hdlcX devices.
> 
> And before anyone asks: Even in 2023 there are real-world deployments of
> those interfaces over Frame Relay in production use.
> 
> This patch doesn't change the default hdlcX netdev MTU, but permits
> userspace to configure a higher MTU, in those cases needed.

I could be reading the code wrong, but:

https://elixir.bootlin.com/linux/latest/source/drivers/net/wan/hdlc.c#L231

static void hdlc_setup_dev(struct net_device *dev)
{
	/* Re-init all variables changed by HDLC protocol drivers,
	 * including ether_setup() called from hdlc_raw_eth.c.
	 */
	dev->flags		 = IFF_POINTOPOINT | IFF_NOARP;
	dev->priv_flags		 = IFF_WAN_HDLC;
	dev->mtu		 = HDLC_MAX_MTU;
	dev->min_mtu		 = 68;
	dev->max_mtu		 = HDLC_MAX_MTU;
	dev->type		 = ARPHRD_RAWHDLC;
	dev->hard_header_len	 = 0;
	dev->needed_headroom	 = 0;
	dev->addr_len		 = 0;
	dev->header_ops		 = &hdlc_null_ops;
}

This change does appear to change dev->mtu, not just dev->max_mtu?
So i'm not sure the commit message is correct?

   Andrew