QUIC - Code to persist if we had talked QUIC to a server and if so what

QUIC - Code to persist if we had talked QUIC to a server and if so what
is the address.

This code is not hooked up into QUIC code yet. Will submit that in
the next CL.

R=rch@chromium.org

Committed: https://crrev.com/1c863aa414026f2ece4ce5b930bcbce1cd07d6f0
Cr-Commit-Position: refs/heads/master@{#296749}

[Ryan Hamilton, 2014-09-25 00:47:30]

I'm not sure I understand. I thought we wanted a single global for this? Or is
the idea that this CL adds the new field that will be associated with the
"dummy" global server name?

[ramant (doing other things), 2014-09-25 02:52:14]

On 2014/09/25 00:47:30, Ryan Hamilton wrote:
> I'm not sure I understand. I thought we wanted a single global for this? Or is
> the idea that this CL adds the new field that will be associated with the
> "dummy" global server name?

This is dummy global server name idea (similar to AlternateProtocolInfo. This is
less risky).

Will have another CL with a global dictionary that stores all globals. May be we
should hold off on this CL. Will work on the globals CL (we could consider this
as backup).

thanks
raman

[Ryan Hamilton, 2014-09-25 02:59:16]

On 2014/09/25 02:52:14, ramant wrote:
> On 2014/09/25 00:47:30, Ryan Hamilton wrote:
> > I'm not sure I understand. I thought we wanted a single global for this? Or
is
> > the idea that this CL adds the new field that will be associated with the
> > "dummy" global server name?
> 
> This is dummy global server name idea (similar to AlternateProtocolInfo. This
is
> less risky).

Ah, ok. And so the QuicStreamFactory will call into this. Are you thinking that
the QuicStreamFactory will define kDummyHostname?

[Ryan Hamilton, 2014-09-25 03:02:57]

Ok, I see what you're doing. This seems like a fine way to get this
functionality in ASAP. We can clean up the interface afterwards.

lgtm

[ramant (doing other things), 2014-09-25 04:14:12]

On 2014/09/25 02:59:16, Ryan Hamilton wrote:
> On 2014/09/25 02:52:14, ramant wrote:
> > On 2014/09/25 00:47:30, Ryan Hamilton wrote:
> > > I'm not sure I understand. I thought we wanted a single global for this?
Or
> is
> > > the idea that this CL adds the new field that will be associated with the
> > > "dummy" global server name?
> > 
> > This is dummy global server name idea (similar to AlternateProtocolInfo.
This
> is
> > less risky).
> 
> Ah, ok. And so the QuicStreamFactory will call into this. Are you thinking
that
> the QuicStreamFactory will define kDummyHostname?

In https://codereview.chromium.org/602763004/ (it is not ready for review yet.
wanted
to talk to you about the IP Address to be saved), I was doing the above.

Will submit these CL's tomorrow after talking with you.

thanks
raman

[ramant (doing other things), 2014-09-25 18:29:32]

The CQ bit was checked by rtenneti@chromium.org

[commit-bot: I haz the power, 2014-09-25 18:29:58]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/605563003/1

[commit-bot: I haz the power, 2014-09-25 18:39:52]

Committed patchset #1 (id:1) as 30107ac5fd05e51f6ea4c7e27c6d2520928baeef

[commit-bot: I haz the power, 2014-09-25 18:42:19]

Patchset 1 (id:??) landed as
https://crrev.com/1c863aa414026f2ece4ce5b930bcbce1cd07d6f0
Cr-Commit-Position: refs/heads/master@{#296749}