qemu-devel
[Top][All Lists]
Advanced

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

Re: [Qemu-devel] [PATCH 16/16] qapi-schema, qemu-options & slirp: Adding


From: Eric Blake
Subject: Re: [Qemu-devel] [PATCH 16/16] qapi-schema, qemu-options & slirp: Adding Qemu options for IPv6 addresses
Date: Tue, 22 Oct 2013 11:52:11 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.0

On 10/22/2013 11:48 AM, Eric Blake wrote:
> On 10/22/2013 11:45 AM, Samuel Thibault wrote:
> 
>>>> It makes sense, yes, the syntax would be [ip6-prefix=net[/netsize]] . I
>>>> just don't know how to express that in qapi :)
>>>
>>> '*ip6_net':'str', '*ip6_netsize':'int'
>>>
>>> passed on the wire as:
>>>
>>> "ip6_net":"fc00::1", "ip6_netsize":64
>>
>> Err, so that means that the user would have to pass
>> "ip6-net=fc00::1,ip6_netsize=64"?  This would look awkward, compared to
>> the ipv4 situation, which is a mere "net=10.0.2.2/24".
> 
> Err, yes.  QMP is machine-parseable.  HMP can let the user abbreviate to
> net=10.0.2.2/24, but QMP should not require reparsing.  So what if it's
> longer, it's easier for machines to manipulate if it's already broken
> into parts.  Trust me - I wouldn't be asking you to split it into two
> fields unless I thought that libvirt would much rather send it to you as
> two fields.

The command line, just like HMP, can use shorthand for convenience.  I
guess you do have a point about consistency of ipv4's @net requiring
post-processing, but that doesn't mean we have to repeat the mistake for
IPv6.

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

Attachment: signature.asc
Description: OpenPGP digital signature


reply via email to

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