Remove indirection for SiteInstance and RenderProcessHost from RenderFrameHost.

Remove indirection for SiteInstance and RenderProcessHost from RenderFrameHost.

Currently the SiteInstance and RenderProcessHost for a RenderFrameHost are retrieved from the associated RenderViewHost. The RFH code is sufficiently along the way to keep track of these on its own.

BUG=304341
Committed: https://crrev.com/9a45a9a69aac0a195da2e5c288c7b4d6f44ea032
Cr-Commit-Position: refs/heads/master@{#313383}

[nasko, 2015-01-27 02:00:13]

nasko@chromium.org changed reviewers:
+ creis@chromium.org

[nasko, 2015-01-27 02:00:14]

Hey Charlie,
This is a small cleanup CL. It teaches RFH about its SiteInstance and
RenderProcessHost instead of asking the RVH for those.
Can you review it for me?
Thanks in advance!
Nasko

[Charlie Reis, 2015-01-27 02:09:04]

Nice-- this is a good step towards eliminating RVH.

https://codereview.chromium.org/881683002/diff/1/content/browser/frame_host/r...
File content/browser/frame_host/render_frame_host_impl.cc (right):

https://codereview.chromium.org/881683002/diff/1/content/browser/frame_host/r...
content/browser/frame_host/render_frame_host_impl.cc:131:
static_cast<SiteInstanceImpl*>(render_view_host->GetSiteInstance())),
Let's pass the SiteInstance as a parameter rather than grabbing it from the RVH.

https://codereview.chromium.org/881683002/diff/1/content/browser/frame_host/r...
File content/browser/frame_host/render_frame_host_impl.h (right):

https://codereview.chromium.org/881683002/diff/1/content/browser/frame_host/r...
content/browser/frame_host/render_frame_host_impl.h:546: // Use a raw pointer to
avoid unnecessary process creation.
Use a raw pointer -> Cache a pointer

[nasko, 2015-01-27 22:30:34]

https://codereview.chromium.org/881683002/diff/1/content/browser/frame_host/r...
File content/browser/frame_host/render_frame_host_impl.cc (right):

https://codereview.chromium.org/881683002/diff/1/content/browser/frame_host/r...
content/browser/frame_host/render_frame_host_impl.cc:131:
static_cast<SiteInstanceImpl*>(render_view_host->GetSiteInstance())),
On 2015/01/27 02:09:04, Charlie Reis (slow until 1-29) wrote:
> Let's pass the SiteInstance as a parameter rather than grabbing it from the
RVH.

Done.

https://codereview.chromium.org/881683002/diff/1/content/browser/frame_host/r...
File content/browser/frame_host/render_frame_host_impl.h (right):

https://codereview.chromium.org/881683002/diff/1/content/browser/frame_host/r...
content/browser/frame_host/render_frame_host_impl.h:546: // Use a raw pointer to
avoid unnecessary process creation.
On 2015/01/27 02:09:04, Charlie Reis (slow until 1-29) wrote:
> Use a raw pointer -> Cache a pointer

Done.

[Charlie Reis, 2015-01-27 22:35:05]

LGTM

[nasko, 2015-01-27 22:38:26]

The CQ bit was checked by nasko@chromium.org

[commit-bot: I haz the power, 2015-01-27 22:39:20]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/881683002/40001

[commit-bot: I haz the power, 2015-01-27 22:43:41]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-01-27 22:44:39]

Patchset 3 (id:??) landed as
https://crrev.com/9a45a9a69aac0a195da2e5c288c7b4d6f44ea032
Cr-Commit-Position: refs/heads/master@{#313383}