On Wed, Feb 10, 2021 at 1:31 AM Daniel P. Berrangé <berra...@redhat.com> wrote:
> On Tue, Feb 09, 2021 at 06:16:57PM -0800, Doug Evans wrote: > > On Thu, Feb 4, 2021 at 10:25 AM Doug Evans <d...@google.com> wrote: > > > > > On Thu, Feb 4, 2021 at 2:03 AM Daniel P. Berrangé <berra...@redhat.com > > > > > wrote: > > > > > >> On Wed, Feb 03, 2021 at 03:35:36PM -0800, dje--- via wrote: > > >> > Add support for ipv6 host forwarding > > >> > > > >> > This patchset takes the original patch from Maxim, > > >> > https://www.mail-archive.com/qemu-devel@nongnu.org/msg569573.html > > >> > and updates it. > > >> > > > >> > New option: -ipv6-hostfwd > > >> > > > >> > New commands: ipv6_hostfwd_add, ipv6_hostfwd_remove > > >> > > > >> > These are the ipv6 equivalents of their ipv4 counterparts. > > >> > > >> Before I noticed this v3, I send a reply to your v2 sugesting > > >> that we don't need to add any new commands/options. We can > > >> use existing inet_parse() helper function to parse the address > > >> info and transparently support IPv4/6 in the existing commands > > >> and options. This matches normal practice elsewhere in QEMU > > >> for IP dual stack. > > >> > > > > > > I'm all for this, fwiw. > > > > > > > > > I should say I'm all for not adding new commands/options. > > Looking at inet_parse() it cannot be used as-is. > > The question then becomes: Will refactoring it buy enough? > > What's the problem your hitting with inet_parse ? > First, this is the inet_parse() function we're talking about, right? int inet_parse(InetSocketAddress *addr, const char *str, Error **errp) https://gitlab.com/qemu-project/qemu/-/blob/master/util/qemu-sockets.c#L618
