Exclude host-side code from the NaCl IRT proxy build.

Exclude host-side code from the NaCl IRT proxy build.
BUG=116317
TEST=compiles
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=167290

[bbudge, 2012-10-11 01:20:31]

This reduces each IRT file by 2-300K.

[brettw, 2012-10-11 18:05:50]

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/interface_list.cc
File ppapi/proxy/interface_list.cc (right):

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/interface_list.cc#...
ppapi/proxy/interface_list.cc:268: AddProxy(API_ID_PPP_INSTANCE,
&ProxyFactory<PPP_Instance_Proxy>);
I would have thought this one and the next line was necessary to register the
message handlers in the plugin side for this interface. Maybe I'm wrong though.
Do these work?

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/interface_list.cc#...
ppapi/proxy/interface_list.cc:281: AddPPP(PPP_InputEvent_Proxy::GetInfo());
Linewise, all of these 4 lines seem necessary.

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/ppb_file_io_proxy.cc
File ppapi/proxy/ppb_file_io_proxy.cc (right):

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/ppb_file_io_proxy....
ppapi/proxy/ppb_file_io_proxy.cc:378: void
PPB_FileIO_Proxy::GeneralCallbackCompleteInHost(
I like to keep the header and .cc ordering the same. And I like to keep the IPC
message handlers separate from the other stuff. Maybe for these cases we should
just have two blocks of ifdefs rather than trying to move some of the host
functions to the middle of the IPC handlers.

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/ppb_graphics_2d_pr...
File ppapi/proxy/ppb_graphics_2d_proxy.cc (right):

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/ppb_graphics_2d_pr...
ppapi/proxy/ppb_graphics_2d_proxy.cc:296: void
PPB_Graphics2D_Proxy::SendFlushACKToPlugin(
Can you revert the change to Graphics2D? Victor is refactoring which will solve
this problem better, and this will cause him some merge conflicts.

[bbudge, 2012-10-11 18:46:02]

I scaled this back as you suggested. It's kind of an odd combination of PPB and
PPP things. I could split it in two. I may do a few more of these, so probably
doesn't make much difference either way.

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/interface_list.cc
File ppapi/proxy/interface_list.cc (right):

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/interface_list.cc#...
ppapi/proxy/interface_list.cc:268: AddProxy(API_ID_PPP_INSTANCE,
&ProxyFactory<PPP_Instance_Proxy>);
On 2012/10/11 18:05:50, brettw wrote:
> I would have thought this one and the next line was necessary to register the
> message handlers in the plugin side for this interface. Maybe I'm wrong
though.
> Do these work?

You are correct.

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/interface_list.cc#...
ppapi/proxy/interface_list.cc:281: AddPPP(PPP_InputEvent_Proxy::GetInfo());
On 2012/10/11 18:05:50, brettw wrote:
> Linewise, all of these 4 lines seem necessary.

Done.

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/ppb_file_io_proxy.cc
File ppapi/proxy/ppb_file_io_proxy.cc (right):

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/ppb_file_io_proxy....
ppapi/proxy/ppb_file_io_proxy.cc:378: void
PPB_FileIO_Proxy::GeneralCallbackCompleteInHost(
On 2012/10/11 18:05:50, brettw wrote:
> I like to keep the header and .cc ordering the same. And I like to keep the
IPC
> message handlers separate from the other stuff. Maybe for these cases we
should
> just have two blocks of ifdefs rather than trying to move some of the host
> functions to the middle of the IPC handlers.

Done for all files where methods were moved.

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/ppb_graphics_2d_pr...
File ppapi/proxy/ppb_graphics_2d_proxy.cc (right):

http://codereview.chromium.org/11094060/diff/1/ppapi/proxy/ppb_graphics_2d_pr...
ppapi/proxy/ppb_graphics_2d_proxy.cc:296: void
PPB_Graphics2D_Proxy::SendFlushACKToPlugin(
On 2012/10/11 18:05:50, brettw wrote:
> Can you revert the change to Graphics2D? Victor is refactoring which will
solve
> this problem better, and this will cause him some merge conflicts.

Done.

[brettw, 2012-10-11 21:18:20]

lgtm

http://codereview.chromium.org/11094060/diff/2019/ppapi/proxy/interface_list.cc
File ppapi/proxy/interface_list.cc (right):

http://codereview.chromium.org/11094060/diff/2019/ppapi/proxy/interface_list....
ppapi/proxy/interface_list.cc:268: #if !defined(OS_NACL)
I'd skip this.

[bbudge, 2012-11-12 18:21:13]

Exclude host-side code from the NaCl IRT proxy build.
BUG=116317
TEST=compiles

[bbudge, 2012-11-12 18:23:23]

Exclude host-side code from the NaCl IRT proxy build.
BUG=116317
TEST=compiles

[bbudge, 2012-11-12 18:26:54]

Exclude host-side code from the NaCl IRT proxy build.
BUG=116317
TEST=compiles

[commit-bot: I haz the power, 2012-11-12 23:25:09]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/bbudge@chromium.org/11094060/15006

[Mark Seaborn, 2012-11-13 03:00:34]

FYI, the change you committed contains debugging printfs and doesn't match patch
set 5.

See
https://src.chromium.org/viewvc/chrome/trunk/src/ppapi/proxy/interface_list.c...

[bbudge, 2012-11-13 03:07:46]

On 2012/11/13 03:00:34, Mark Seaborn wrote:
> FYI, the change you committed contains debugging printfs and doesn't match
patch
> set 5.
> 
> See
>
https://src.chromium.org/viewvc/chrome/trunk/src/ppapi/proxy/interface_list.c...

Thanks, the CQ got stalled and I forgot I'd added these. Fixing.