Remove loss detection from QuicConfig now that connection options has

Remove loss detection from QuicConfig now that connection options has
replaced it.

Merge internal change: 75952172

Chromium specific changes:
+ Deleted enable_quic_time_based_loss_detection from 
  NetworkSession params.
+ Deleted enable-quic-time-based-loss-detection and
  disable-quic-time-based-loss-detection command line
  switches.

R=rch@chromium.org, sky@chromium.org

[ramant (doing other things), 2014-09-26 02:54:43]

rtenneti@chromium.org changed reviewers:
+ sky@chromium.org
- thakis@chromium.org, thestig@chromium.org

[Ryan Hamilton, 2014-09-26 03:26:39]

lgtm

It'd be a good idea to also land a finch change so that the existing
EnabledWithTimeBasedLossDetection field trial groups are configured to add TIME
to connection_options

[sky, 2014-09-26 16:03:49]

chrome LGTM

[ramant (doing other things), 2014-09-26 17:27:56]

Patchset #2 (id:20001) has been deleted

[ramant (doing other things), 2014-09-26 17:32:47]

Hi Ryan and sky,
  Added back the WithTimeBasedLossDetection FieldTrial and the command line
options. The only change this CL does for "Chrome" code is to set the kTIME
connection options. Restored the deleted code.

  PTAL.

thanks
raman

[sky, 2014-09-26 19:45:27]

SLGTM

[Ryan Hamilton, 2014-09-26 21:18:51]

On 2014/09/26 17:32:47, ramant wrote:
> Hi Ryan and sky,
>   Added back the WithTimeBasedLossDetection FieldTrial and the command line
> options. The only change this CL does for "Chrome" code is to set the kTIME
> connection options. Restored the deleted code.
> 
>   PTAL.
> 
> thanks
> raman

I thought that the finch change which landed today changed the field trial to
add kTIME to connection_options via the existing connection_options handling
code. In other words, I think you can keep the code removed. But ping me when
you're back and we can chat.

[ramant (doing other things), 2014-09-26 21:57:52]

On 2014/09/26 21:18:51, Ryan Hamilton wrote:
> On 2014/09/26 17:32:47, ramant wrote:
> > Hi Ryan and sky,
> >   Added back the WithTimeBasedLossDetection FieldTrial and the command line
> > options. The only change this CL does for "Chrome" code is to set the kTIME
> > connection options. Restored the deleted code.
> > 
> >   PTAL.
> > 
> > thanks
> > raman
> 
> I thought that the finch change which landed today changed the field trial to
> add kTIME to connection_options via the existing connection_options handling
> code. In other words, I think you can keep the code removed. But ping me when
> you're back and we can chat.

Reverted my changes in Patch Set 2. Patch Set 3 is same as Patch Set 1. IMO,
code review is optional (because Patch Set 1 was already LGTM'ed)

sky@: I thought I need this code similar to "kQuicFieldTrialPacingSuffix". rch
showed me even that code could be deleted. command line option
kQuicConnectionOptions does what enable-...pacing and enable-...loss...
arguments do.

thanks
raman

[Ryan Hamilton, 2014-09-26 22:04:29]

LGTM. The finch changes should work well with this.

[Nico, 2014-10-27 21:26:53]

(do you still need me to do anything here, or can this land?)

[ramant (doing other things), 2014-10-27 21:53:49]

On 2014/10/27 21:26:53, Nico wrote:
> (do you still need me to do anything here, or can this land?)

No need to do anything. sky LGTM'ed and it was landed as part of
https://codereview.chromium.org/605733006/ (a merged CL). Thanks for asking.