Remove a bunch of unused network flags.

Remove a bunch of unused network flags.

--enable-sdch (renamed to --enable-sdch-over-https)
--enable-spdy2
--max-spdy-concurrent-streams (kept the ability in the networking library to expose this for configuration purposes, but chromium will not configure it)
--enable-ip-pooling
--disable-ip-pooling

BUG=344673, 303957, 344684, 344698, 344793
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=254316

[willchan no longer on Chromium, 2014-02-19 00:05:53]

mef: Please review the sdch change
szym: Please review the IPv6 change
rch/johnny: Please review the spdy changes. I'm somewhat tempted to yank out the
HttpNetworkSessions::Param value for max spdy streams and hardcode a constant.
But I kept it in there for now.

[szym, 2014-02-19 00:13:29]

https://codereview.chromium.org/169053004/diff/1/chrome/browser/io_thread.cc
File chrome/browser/io_thread.cc (right):

https://codereview.chromium.org/169053004/diff/1/chrome/browser/io_thread.cc#...
chrome/browser/io_thread.cc:191: if
(command_line.HasSwitch(switches::kDisableIPv6)) {
IMO if anything deserves to be left behind, it is --enable-ipv6.

If Chrome incorrectly disables IPv6, users cannot reach local IPv6-only
resources, so --enable-ipv6 is a last resort workaround.

In contrast, if Chrome incorrectly enables IPv6, it costs a few hundred
milliseconds in performance for every new socket. I don't think people will
bother to use the command-line flag to work around such a performance problem.

[willchan no longer on Chromium, 2014-02-19 00:21:43]

https://codereview.chromium.org/169053004/diff/1/chrome/browser/io_thread.cc
File chrome/browser/io_thread.cc (right):

https://codereview.chromium.org/169053004/diff/1/chrome/browser/io_thread.cc#...
chrome/browser/io_thread.cc:191: if
(command_line.HasSwitch(switches::kDisableIPv6)) {
On 2014/02/19 00:13:29, szym wrote:
> IMO if anything deserves to be left behind, it is --enable-ipv6.
> 
> If Chrome incorrectly disables IPv6, users cannot reach local IPv6-only
> resources, so --enable-ipv6 is a last resort workaround.
> 
> In contrast, if Chrome incorrectly enables IPv6, it costs a few hundred
> milliseconds in performance for every new socket. I don't think people will
> bother to use the command-line flag to work around such a performance problem.

Oh, maybe I got this wrong. Did you think we should move *both* to
net-internals? Or just one and remove the other completely? I'm open to
whatever.

[Ryan Hamilton, 2014-02-19 01:29:18]

lgtm

SPDY changes LGTM.

[Johnny, 2014-02-19 01:37:14]

SPDY lgtm

> rch/johnny: Please review the spdy changes. I'm somewhat tempted to yank out
the
> HttpNetworkSessions::Param value for max spdy streams and hardcode a constant.
> But I kept it in there for now.

I'm in favor of a kOnstant over retaining
globals->max_spdy_concurrent_streams_limit. Up to you.

[szym, 2014-02-19 02:29:16]

https://codereview.chromium.org/169053004/diff/1/chrome/browser/io_thread.cc
File chrome/browser/io_thread.cc (right):

https://codereview.chromium.org/169053004/diff/1/chrome/browser/io_thread.cc#...
chrome/browser/io_thread.cc:191: if
(command_line.HasSwitch(switches::kDisableIPv6)) {
On 2014/02/19 00:21:43, willchan wrote:
> On 2014/02/19 00:13:29, szym wrote:
> > IMO if anything deserves to be left behind, it is --enable-ipv6.
> > 
> > If Chrome incorrectly disables IPv6, users cannot reach local IPv6-only
> > resources, so --enable-ipv6 is a last resort workaround.
> > 
> > In contrast, if Chrome incorrectly enables IPv6, it costs a few hundred
> > milliseconds in performance for every new socket. I don't think people will
> > bother to use the command-line flag to work around such a performance
problem.
> 
> Oh, maybe I got this wrong. Did you think we should move *both* to
> net-internals? Or just one and remove the other completely? I'm open to
> whatever.

In contrast to -async-dns (which can be moved to net-internals), we do not have
a persistent preference (in chrome://flags or group policy) that would allow
users to override this, so I'd prefer to leave --enable-ipv6 command line flag.

[willchan no longer on Chromium, 2014-02-19 07:17:51]

szym: I brought back the IPv6 flags. I think this should be in the clear now for
that. We should continue that discussion, but I'm unblocking this CL.

rch/johnny: I saw the IP pooling flags and killed them off. I think SPDY IP
pooling should always be on, so I just enabled them always.

johnny: I'm leaning towards agreeing with you on the max streams. I'm leaving
soon for vacation, so I may save that for another day.

[Ryan Hamilton, 2014-02-19 16:26:24]

On 2014/02/19 07:17:51, willchan wrote:
> rch/johnny: I saw the IP pooling flags and killed them off. I think SPDY IP
> pooling should always be on, so I just enabled them always.

Sounds good to me.

> johnny: I'm leaning towards agreeing with you on the max streams. I'm leaving
> soon for vacation, so I may save that for another day.

Also sounds good

[mef, 2014-02-24 21:41:37]

https://codereview.chromium.org/169053004/diff/250001/chrome/browser/chrome_b...
File chrome/browser/chrome_browser_main.cc (left):

https://codereview.chromium.org/169053004/diff/250001/chrome/browser/chrome_b...
chrome/browser/chrome_browser_main.cc:1403:
net::SdchManager::EnableSdchSupport(false);
Is it OK that if we remove this option, then there is no way to disable SDCH
support?

https://codereview.chromium.org/169053004/diff/250001/chrome/browser/chrome_b...
File chrome/browser/chrome_browser_main.cc (right):

https://codereview.chromium.org/169053004/diff/250001/chrome/browser/chrome_b...
chrome/browser/chrome_browser_main.cc:1394: if
(parsed_command_line().HasSwitch(switches::kEnableSdchOverHttps)) {
Would it make sense to replace command line switch with finch trial? AFIUI it is
possible to explicitly enable/disable finch trials locally.

[willchan no longer on Chromium, 2014-02-24 21:45:57]

https://codereview.chromium.org/169053004/diff/250001/chrome/browser/chrome_b...
File chrome/browser/chrome_browser_main.cc (left):

https://codereview.chromium.org/169053004/diff/250001/chrome/browser/chrome_b...
chrome/browser/chrome_browser_main.cc:1403:
net::SdchManager::EnableSdchSupport(false);
On 2014/02/24 21:41:37, mef wrote:
> Is it OK that if we remove this option, then there is no way to disable SDCH
> support?

I assert yes, according to the policies laid out in the chromium-dev thread.

https://codereview.chromium.org/169053004/diff/250001/chrome/browser/chrome_b...
File chrome/browser/chrome_browser_main.cc (right):

https://codereview.chromium.org/169053004/diff/250001/chrome/browser/chrome_b...
chrome/browser/chrome_browser_main.cc:1394: if
(parsed_command_line().HasSwitch(switches::kEnableSdchOverHttps)) {
On 2014/02/24 21:41:37, mef wrote:
> Would it make sense to replace command line switch with finch trial? AFIUI it
is
> possible to explicitly enable/disable finch trials locally.

I dunno. It's up to you :)

[willchan no longer on Chromium, 2014-02-27 23:10:40]

Remove the rest of pooling.

[willchan no longer on Chromium, 2014-02-27 23:41:51]

merge

[mef, 2014-02-28 16:55:36]

lgtm

[willchan no longer on Chromium, 2014-02-28 22:43:26]

The CQ bit was checked by willchan@chromium.org

[commit-bot: I haz the power, 2014-02-28 22:44:39]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/willchan@chromium.org/169053004/480001

[commit-bot: I haz the power, 2014-02-28 23:01:51]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-02-28 23:01:51]

Retried try job too often on chromium_presubmit for step(s) presubmit
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=chromium_p...

[willchan no longer on Chromium, 2014-02-28 23:06:49]

Oops, I missed a chrome/ OWNER. +sky

[sky, 2014-02-28 23:27:24]

LGTM

[willchan no longer on Chromium, 2014-02-28 23:34:17]

The CQ bit was checked by willchan@chromium.org

[commit-bot: I haz the power, 2014-02-28 23:35:34]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/willchan@chromium.org/169053004/480001

[commit-bot: I haz the power, 2014-03-01 01:08:17]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/willchan@chromium.org/169053004/480001

[commit-bot: I haz the power, 2014-03-01 01:29:46]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/willchan@chromium.org/169053004/480001

[commit-bot: I haz the power, 2014-03-01 01:42:05]

Change committed as 254316