PPAPI: Make PPP_MessageHandler work in PNaCl

PPAPI: Make PPP_MessageHandler work in PNaCl

The PNaCl ABI does not work in general with passing structs by-value.
Rather than add more complication to the shim, this changes
HandleBlockingMessage's parameter to a const-pointer param, and the return
to an out-param by pointer.

Note I wanted to pass PP_Var by const-pointer, while its "default" is
by value. So I added a new "constptr_in" param style.
(Maybe I should invert that and add a byvalue instead?)

Verified backwards compat with manual testing.

BUG=384539
R=asvitkine@chromium.org, binji@chromium.org, teravest@chromium.org

Committed: https://crrev.com/970fa0cfd701760b7dda55de4233d614646d5c71
Cr-Commit-Position: refs/heads/master@{#295341}

[dmichael (off chromium), 2014-09-10 23:08:45]

dmichael@chromium.org changed reviewers:
+ mseaborn@chromium.org, teravest@chromium.org

[dmichael (off chromium), 2014-09-10 23:08:46]

teravest: Primary
mseaborn: In case you want to look at the PNaCl-relevant bits (it's just what we
already talked about).

[teravest, 2014-09-11 15:42:41]

lgtm

This seems pretty straightforward to me.

Are the current users aware that version 0.1 is going to go away in the next
milestone of Chrome? It's how dev channel APIs are meant to work, but I don't
know if they're aware of that.

https://codereview.chromium.org/564573002/diff/1/ppapi/proxy/message_handler.cc
File ppapi/proxy/message_handler.cc (right):

https://codereview.chromium.org/564573002/diff/1/ppapi/proxy/message_handler....
ppapi/proxy/message_handler.cc:28: const_cast<const
PP_Var*>(&message_data.get()));
As discussed off-thread, the const_cast<> here looks weird, but I don't think
you need to make any changes to CallWhileUnlocked to handle that better in this
change.

[dmichael (off chromium), 2014-09-16 21:19:36]

Patchset #2 (id:20001) has been deleted

[dmichael (off chromium), 2014-09-16 21:20:16]

Sorry to do this to you, but this is much bigger now with
backwards-compatibility

[teravest, 2014-09-17 14:59:29]

On 2014/09/16 21:20:16, dmichael wrote:
> Sorry to do this to you, but this is much bigger now with
> backwards-compatibility

lgtm

[dmichael (off chromium), 2014-09-17 15:15:22]

dmichael@chromium.org changed reviewers:
+ asvitkine@chromium.org, binji@chromium.org

[dmichael (off chromium), 2014-09-17 15:15:23]

+asvitkine for histograms, binji for nacl_sdk

[Alexei Svitkine (slow), 2014-09-17 15:20:46]

lgtm

[binji, 2014-09-17 16:44:46]

native_client_sdk lgtm

[dmichael (off chromium), 2014-09-17 17:26:27]

The CQ bit was checked by dmichael@chromium.org

[commit-bot: I haz the power, 2014-09-17 17:27:43]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patchset/564573002/100001

[commit-bot: I haz the power, 2014-09-17 17:40:35]

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

[commit-bot: I haz the power, 2014-09-17 17:40:38]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[dmichael (off chromium), 2014-09-17 20:58:53]

Committed patchset #5 (id:100001) manually as 970fa0c (presubmit successful).

[commit-bot: I haz the power, 2014-09-17 20:58:53]

Patchset 5 (id:??) landed as
https://crrev.com/970fa0cfd701760b7dda55de4233d614646d5c71
Cr-Commit-Position: refs/heads/master@{#295341}