Revert 194613 "Pepper: Autogenerate thunk for PPB_URL_Loader."

Revert 194613 "Pepper: Autogenerate thunk for PPB_URL_Loader."

> Pepper: Autogenerate thunk for PPB_URL_Loader.
> 
> This requires another annotation:
>   "always_set_output_parameters"
> 
> This annotation ensures that the thunk implementation will zero-out the value of all output parameters on failure.
> 
> Tested:
>   Built chrome and browser_tests.
> 
> BUG=
> 
> Review URL: https://codereview.chromium.org/14007010

TBR=teravest@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=194617

[asargent_no_longer_on_chrome, 2013-04-17 17:05:02]

Committed patchset #1 manually as r194617.

[teravest, 2013-04-17 17:07:10]

lgtm, I suppose. Did I miss a build breakage?

On Wed, Apr 17, 2013 at 11:05 AM,  <asargent@chromium.org> wrote:
> Committed patchset #1 manually as r194617.
>
> https://codereview.chromium.org/14195008/

[asargent_no_longer_on_chrome, 2013-04-17 17:09:42]

Looks like it was clang warnings. See the buildbot failure email that was
sent - it links to:

http://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%2...

which shows:

ppapi/thunk/ppb_gamepad_thunk.cc:30:26:error: 'memset' call operates
on objects of type 'struct PP_GamepadsSampleData' while the size is
based on a different type 'struct PP_GamepadsSampleData *'
[-Werror,-Wsizeof-pointer-memaccess]
  memset(data, 0, sizeof(data));
         ~~~~            ^~~~
ppapi/thunk/ppb_gamepad_thunk.cc:30:26: note: did you mean to
dereference the argument to 'sizeof' (and multiply it by the number of
elements)?
  memset(data, 0, sizeof(data));
                         ^~~~
1 error generated.make: ***
[out/Debug/obj.target/ppapi_shared/ppapi/thunk/ppb_gamepad_thunk.o]
Error 1
ppapi/thunk/ppb_url_loader_thunk.cc:63:34:error: 'memset' call
operates on objects of type 'int64_t' (aka 'long long') while the size
is based on a different type 'int64_t *' (aka 'long long *')
[-Werror,-Wsizeof-pointer-memaccess]
    memset(bytes_sent, 0, sizeof(bytes_sent));
           ~~~~~~~~~~            ^~~~~~~~~~
ppapi/thunk/ppb_url_loader_thunk.cc:63:34: note: did you mean to
dereference the argument to 'sizeof' (and multiply it by the number of
elements)?
    memset(bytes_sent, 0, sizeof(bytes_sent));
                                 ^~~~~~~~~~
ppapi/thunk/ppb_url_loader_thunk.cc:64:46:error: 'memset' call
operates on objects of type 'int64_t' (aka 'long long') while the size
is based on a different type 'int64_t *' (aka 'long long *')
[-Werror,-Wsizeof-pointer-memaccess]
    memset(total_bytes_to_be_sent, 0, sizeof(total_bytes_to_be_sent));
           ~~~~~~~~~~~~~~~~~~~~~~            ^~~~~~~~~~~~~~~~~~~~~~
ppapi/thunk/ppb_url_loader_thunk.cc:64:46: note: did you mean to
dereference the argument to 'sizeof' (and multiply it by the number of
elements)?
    memset(total_bytes_to_be_sent, 0, sizeof(total_bytes_to_be_sent));
                                             ^~~~~~~~~~~~~~~~~~~~~~
ppapi/thunk/ppb_url_loader_thunk.cc:76:38:error: 'memset' call
operates on objects of type 'int64_t' (aka 'long long') while the size
is based on a different type 'int64_t *' (aka 'long long *')
[-Werror,-Wsizeof-pointer-memaccess]
    memset(bytes_received, 0, sizeof(bytes_received));
           ~~~~~~~~~~~~~~            ^~~~~~~~~~~~~~
ppapi/thunk/ppb_url_loader_thunk.cc:76:38: note: did you mean to
dereference the argument to 'sizeof' (and multiply it by the number of
elements)?
    memset(bytes_received, 0, sizeof(bytes_received));
                                     ^~~~~~~~~~~~~~
ppapi/thunk/ppb_url_loader_thunk.cc:77:50:error: 'memset' call
operates on objects of type 'int64_t' (aka 'long long') while the size
is based on a different type 'int64_t *' (aka 'long long *')
[-Werror,-Wsizeof-pointer-memaccess]
    memset(total_bytes_to_be_received, 0, sizeof(total_bytes_to_be_received));
           ~~~~~~~~~~~~~~~~~~~~~~~~~~            ^~~~~~~~~~~~~~~~~~~~~~~~~~
ppapi/thunk/ppb_url_loader_thunk.cc:77:50: note: did you mean to
dereference the argument to 'sizeof' (and multiply it by the number of
elements)?
    memset(total_bytes_to_be_received, 0, sizeof(total_bytes_to_be_received));
                                                 ^~~~~~~~~~~~~~~~~~~~~~~~~~





On Wed, Apr 17, 2013 at 10:06 AM, Justin TerAvest <teravest@chromium.org>wrote:

> lgtm, I suppose. Did I miss a build breakage?
>
> On Wed, Apr 17, 2013 at 11:05 AM,  <asargent@chromium.org> wrote:
> > Committed patchset #1 manually as r194617.
> >
> > https://codereview.chromium.org/14195008/
>