qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] Contribution - L2TPv3 transport


From: Anton Ivanov (antivano)
Subject: Re: [Qemu-devel] Contribution - L2TPv3 transport
Date: Tue, 4 Mar 2014 15:22:17 +0000
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20131103 Icedove/17.0.10

Apologies, missed to diff the json definitions.

Attached.

On 04/03/14 15:19, Anton Ivanov wrote:
> Attached is a revised version.
>
>     * I am still keeping the sendmsg instead of iov_send because I 
> have not had the time to review udp.c in the kernel and do the 
> relevant regression testing to see if they connect() still breaks 
> bind() on multihomed hosts as it did in 2.6. We can revisit that at a 
> later date, without doing the proper investigation I am not 
> comfortable trying this.
>
>     * I have killed completely parse6 and replaced that by a common 
> getaddrinfo()  with a selectable address family based on the boolean 
> flags. This also allows to force a specific v4 or v6 address choice 
> for dst.
>
>     * addresses now are specified separately from ports and ports are 
> deliberately strings so you can specify them as a protocol.
>
>     * mode bitmask is gone it is all booleans now - both for option 
> invocation and internally in the code.
>
>     * I have added the extra offset back in so it is feature by 
> feature compatible with linux kernel implementation.
>
>     * All mallocs are now to exact size and
>
>     * Indentation, style, debug, etc are all as requested now.
>
>     * l2tpv3.h has become surplus to requirements and is gone now.
>
>     * it now has proper cleanup.
>
>     * I have tested it for a few setups (mostly v4) and it works as 
> advertised. I need to rewrite my tests scripts for the new options 
> names to give it a full test. I would not expect that to show any 
> problems though - core send/receive logic is unchanged from the 
> original version.
>
>     * I have not yet addressed the page size and page alignment of 
> buffers items - leaving that open for an RFC on how to get the max 
> performance there and how to make it expandable lately so one can 
> re-configure it for large MTU/jumbo frames.
>
> Example magic incantation to invoke it with the new options:
>
> #!/bin/sh
>
> kvm -hda kvm.img  -m 1024 \
>    -net nic,vlan=0,model=virtio,macaddr=0a:98:fc:96:83:01 \
>    -net 
> l2tpv3,vlan=0,udp,src=192.168.63.1,srcport=1700,dst=192.168.63.1,dstport=1701,cookie64,txcookie=0x0123456789abcdef,rxcookie=0xfedcba9876543210,rxsession=0xffffffff,txsession=0xffffffff,counter
>
> A.

Attachment: l2tpv3-json.diff
Description: l2tpv3-json.diff


reply via email to

[Prev in Thread] Current Thread [Next in Thread]