Initial support for HTTP/2 dependencies and weights

Initial support for HTTP/2 dependencies and weights.

A non-exclusive root dependency is always emitted, and parsed
dependencies are ignored. Weights are mapped to existing SPDY
priorities, and vice versa.

BUG=345769
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=271555

[Johnny, 2014-05-19 21:14:46]

Last blocker for h2-12 testing in Chromium. Also under concurrent review
server-side, but I'm mailing this out now to unblock interop testing. I'll deal
with any mismatch later.

[Ryan Hamilton, 2014-05-19 22:56:41]

LGTM, though consider using constants.

https://codereview.chromium.org/287003011/diff/40001/net/spdy/spdy_framer.cc
File net/spdy/spdy_framer.cc (right):

https://codereview.chromium.org/287003011/diff/40001/net/spdy/spdy_framer.cc#...
net/spdy/spdy_framer.cc:188: return GetControlFrameHeaderSize() + 5;
I'm not terribly in love with the magic number here. Any chance you can extract
this to a constant? Perhaps that's not super important, though I do think it'd
be a bit more readable.

Perhaps it should be 2 constants, actually... one for weight size, and one for
dependency size?

[Johnny, 2014-05-19 23:27:17]

https://codereview.chromium.org/287003011/diff/40001/net/spdy/spdy_framer.cc
File net/spdy/spdy_framer.cc (right):

https://codereview.chromium.org/287003011/diff/40001/net/spdy/spdy_framer.cc#...
net/spdy/spdy_framer.cc:188: return GetControlFrameHeaderSize() + 5;
On 2014/05/19 22:56:41, Ryan Hamilton wrote:
> I'm not terribly in love with the magic number here. Any chance you can
extract
> this to a constant? Perhaps that's not super important, though I do think it'd
> be a bit more readable.
> 
> Perhaps it should be 2 constants, actually... one for weight size, and one for
> dependency size?

Done. Though, now this particularly calls attention to all of the /other/ places
that should do this in spdy_framer.cc :)

[Johnny, 2014-05-19 23:38:26]

The CQ bit was checked by jgraettinger@chromium.org

[commit-bot: I haz the power, 2014-05-19 23:38:47]

CQ is trying da patch. Follow status at

https://chromium-status.appspot.com/cq/jgraettinger@chromium.org/287003011/10...

[commit-bot: I haz the power, 2014-05-20 03:05:13]

Change committed as 271555